• Sand bedding was cleaned 3 times per day (during milking) to remove all organic materials. In general, fresh new sand was added to the bedding once per week, according to the bedding thickness
• Mean herd cumulative incidence of clinical mastitis (CICM) was 3.3% per month in the herds and median CICM was 3.0% per month (range = 1.7 to 8.1 per herd)
• coli was most frequently isolated (14.4%), followed by Klebsiella spp. (13.0%), CNS (11.3%), Strep. dysgalactiae (10.5%), Staph. aureus (10.2%), and other streptococci (8.0%). Enterobacter spp.,Strep. agalactiae, and Strep. uberis were isolated in 5.5, 2.8, and 2.1% of samples, respectively
• Of 3,288 Clinical Mastitis (CM) samples, 1,750 were collected from herds using sand bedding, whereas 1,538 samples were from herds using organic bedding materials
• Mean monthly CICM was 2.9 and 3.7% for herds using sand and organic bedding material, respectively. Streptococcus dysgalactiae was more frequently isolated in herds using sand bedding than those using organic bedding, whereas Klebsiella spp. and other streptococci were more prevalent in herds using organic bedding.

• Restricted to farms in China
• Much variation in management possible between farms and over time.
• Mastitis outcome can be influenced by many other factors than bedding, so trends can be reported and associations but solid conclusions are tentative
• There are many seasonal and geographical environment factors that vary across such a wide region as China that may also affect mastitis outcomes
• Culturing on agar may not always show growth of causative organisms
• No mastitis samples were recorded in February due to the Spring festival

• The proportion of udders that were classified as clean (score 1 or 2) was 68%, 82%, 54%, and 95% for cows housed in pens containing NES, RS, SBMS, and DBMS, respectively
• No association was found between HK score and teat skin bacterial count.
• Higher hygiene scores (dirtier teats) were associated with higher bacterial counts
• Bacterial counts of teat skin swabs from front teats of cows in pens containing RS and SBMS were significantly less than those of rear teats of cows in pens containing DBMS or NES

• Low population numbers
• Mostly from primiparous cows
• Manual observations may be subjective

• Questions related to 7 major areas: sociodemographic and farm characteristics, milking proficiency, milking systems, cow environment, infected cow monitoring and treatment, farm labor, and attitudes toward mastitis and related antimicrobial use
• Amount of antimicrobial drug use for clinical mastitis by intramammary or systemic administration

• Use of sand or mattresses for bedding, rather than older styles of housing, were associated with decreased systemic antimicrobial use but also there is an association of less systemic use on higher managed farm types

• Survey was only to Grade A certified dairy farms, so may not be representative of the whole dairy population
• Responses were dependent on farmer opinion and self-reported frequency of antimicrobial use
• There was a huge amount of variation in management practices across the surveyed population
• Bedding use was only a very small part of the survey and was based on Sand or Mattress (not specific other bedding types)
• Bedding associations based on multivariate regression analysis on low-level evidence foundation (survey and self opinion)

• Freestall housed cows lay down for 2 hours longer during far-off and close-up periods (p < 0.01; DF: 76), but no differences occurred during calving and postpartum compared to pastured cows
• No difference in bacteria cultured from either group

• Low cow numbers
• Overproportion of primiparous cows (>60%)
• Only studied in Autumn/early winter
• Feedspace per cow was 60cm in headlocks for indoor and free in pasture (lower than recommended 75cm in literature and possible for bullying in pasture system, potentially reducing feed intake)
• Comparing sand and pasture, not bedding types in same system, so confusing results for general indoor management and pasture
• Manual observations may be subjective

• Bacterial culture of bedding
• Teat swabs from one teat per cow from each cow, once a week, pre- and post-preparation for milking
• Linear regression modelling

• The median number of total Gram-negative bacteria in DBMS was 13 times as great as the median number in SBMS and 60 and 216 times as great as the median numbers in RS and NES
• Not much difference between recycled and new sand in numbers of coliforms
• Numbers of total gram-negative bacteria, coliforms, and Klebsiella were significantly greater in the summer than in other seasons
• 12 times as many streptococci in shallow-bedded manure solids as in recycled sand
• Across all bacteria and bedding types, the mean number of bacteria recovered from pre-milking swabs (CFU/swab) were approximately 2 to 3 logs less than the corresponding number of bacteria in bedding
• 4, 8, and 19 times as many total gram-negative bacteria recovered from cows bedded on SBMS, RS, and DBMS versus NS.
• Post-prep swabs revealed less bacteria in all spp. and all bedding types
• coliform bacteria (Escherichia spp., Klebsiella spp., and Enterobacter spp.) were only 2 to 7% of total gram-negative bacteria in bedding

• Primarily primiparous cows (96 vs 15 multiparous)
• Other cows entered the pens when cows ended their lactations

• Incidence of subclinical mastitis (SM): >200k cells per quarter, sampled every 28 days from every quarter
• Incidence of clinical mastitis (CM): abnormal milk appearance. All affected cases were cultured on agar

• The proportion of SM samples that were culture negative was numerically greater in quarters of primiparous cows bedded with NES and DBMS as compared with those bedded with RS and SBMS
• Far fewer cases of CM were caused by Gm –ve bacteria for primiparous cows housed in pens containing NES than in other bedding types
• Majority of SM cases were cultured negative (71%), meaning that there was no association with cultured species and bedding type for SM.
• Almost half of milk samples from CM cases did not result in the identification of a pathogen
• The overall incidence of CM was 0.26 cases/1,000 and was not associated with bedding types
• There was a tendency for longer survival times to first case of CM for quarters of cows in pens containing NES as compared with quarters of cows bedded with RS (P = 0.056) or DBMS (P = 0.086)

• Primarily primiparous cows (96 vs 15 multiparous)
• Other cows entered the pens when cows ended their lactations
• Time in pen for RS and DBMS was significantly longer than NS and SBMS groups

19mm thick rubber mattresses in freestalls or 30cm deep bedded sand. Each cow spent 24 hours on one type and then 24 hours on the other. Then they had access to both surfaces for a further 6 days and lameness scored at the end. Concrete slatted flooring in aisles.

Scales were the experimenters’ own and defined in the methods section of the paper:

• Lameness on 1-5 scale
• Hoof lesions from trimming and manual ID
• Lying times and frequency of changing position
• CCTV and manual observation

• For cows kept on deep-bedded sand, the total lying time, the number of cows observed to lie in the lateral position, and the frequency of lying bouts were higher than when kept on the rubber mats
• On sand, the lying time ranged from 28 min to 16 h, and while on rubber, it ranged from 6 min to 14 h
• The number of cows performing interruptions of lying down did not differ between the 2 surfaces, but while kept on sand, the cows lay down and got up faster than the rubber mats

• Only 4 months of study and prior to this, cows were on loose housing. As lameness lesions take time to manifest clinically, this study cannot imply that bedding types reduce onset of lameness
• Cows treated with medication including NSAID were excluded
• Cows with sole haemorrhage and no other lesion were excluded
• Cows that improved to lameness score 1 or 2 were excluded from the trial (10 cows in total excluded)
• Manual observations may be subjective

Lying behavior modelled using variables: Bedding type, quantity (< or > 2cm depth) and cleanliness

• Automated accelerometer readings
• Lying times
• Lameness on 1-5 scale
• Number of hock and knee injuries

• Mean herd-level daily lying time ranged from 8.2 to 13.2 h/d and individual daily lying time for cows ranged from 1.3 to 22.1 h/d.
• Daily lying time was associated with the same risk factors as the other measures of lying behavior
• Bout duration was shorter for cows with injured hocks
• At the herd level, cows housed in stalls with sand had an increased average daily lying time of 1.44 h/d compared with cows housed in stalls with wood shavings

• Logistic regression and multivariate modelling on many farm and management factors
• Farms had to have >=7000L/lactation average so may not be representative of all farms in country
• Lameness by manual assessment
• Manual observations may be subjective

• Gram-negative bacteria, coliform, and streptococcal counts in RMS windrows were reduced after 4 wk composting compared with counts in windrows before composting.
• Klebsiella counts and DM did not differ
• Composting recycled manure solids (RMS) offered minimal advantages over fresh RMS in reducing bacterial counts of common mastitis pathogens in bedding for dairy cows in freestalls

• Low cow numbers
• Multiparous cows only
• Mid-late lactation cows (most studies report more lameness in early to mid lactation)

19mm thick rubber mattresses in freestalls or 30cm deep bedded sand. Each cow spent 24 hours on one type and then 24 hours on the other. Then they had access to both surfaces for a further 6 days and lameness scored at the end.

Concrete slatted flooring in aisles

• Scales were Lameness on 1-5 scale (according to Thomsen P.T., Munksgaard L., Togersen F.A., (2008), Evaluation of a lameness scoring system for dairy cows, Journal of Dairy Science, 91, 116-126)
• Hoof lesions from trimming and manual ID
• Lying times and frequency of changing position
• CCTV and manual observation

• The cows spent more time on the deep-bedded sand than on the rubber surface.
• On sand, cows spent the majority of the time lying, whereas on the rubber surface, they spent most of the time upright

• Only 4 months of study and prior to this, cows were on loose housing. As lameness lesions take time to manifest clinically, this study cannot imply that bedding types reduce onset of lameness
• Cows treated with medication including NSAID were excluded
• Cows with sole haemorrhage and no other lesion were excluded
• Cows that improved to lameness score 1 or 2 were excluded from the trial (10 cows in total excluded)
• Manual observations may be subjective

• Of the total bacteria sampled coliform, streptococci, staphylococci, and Bacillus comprised 1.86%, 20.61%, 52.28%, and 25.25% of all bacteria, respectively
• Coliforms and staphylococci measured a strong correlation with composite temperature (increasing with increasing temp.) and negatively with moisture
• Managing the CBP system for moisture, temperature, C:N ratio, and space per cow may help to reduce some bacterial species counts, but the bacterial load in the bedding will likely remain high.

• Many different stocking densities and management conditions between barns, especially concerning hygiene
• 32 barns were for lactating cows and 13 for hospital pens, so comparing healthy with sick cows may bring confounding variables

Lying times and lameness scores (1-5) correlated to:

management measures, including stall design, bedding, and flooring type within the pen

• Daily lying time decreased with increasing barn age and increased with the use of deep and sand bedding
• Daily lying time decreased as stall stocking density increased, Daily lying time increased in the presence of rubber flooring in part of the pen, pen space per cow, and the percentage of stalls with fecal contamination,
• Frequency of lying bouts decreased and average bout duration increased with the use of deep and sand bedding

• All farms in the study were from a quality-assurance management program, of which 40 cows were chosen from each farm. As these were already on an assurance scheme, they may not be representative of USA population of dairy farms/cows.
• Surveys
• Retrospective study
• Manual observation of lameness
• USA data, not Canadian
• Data collected in selected months (Mar-May and Jul-Oct 2010)
• Manual observations may be subjective

Four bedding types were tested: kiln-dried spruce shavings, “green” (not kiln dried) hemlock sawdust, processed manure-pack, and sand as well as broth control.

3 samples taken per type.

Growth of Prototheca

In vitro study

• Bedding type had a significant effect on Prototheca growth in vitro, and this effect was associated with increased growth in manure, sawdust, and sand beddings.
• Kiln-dried spruce shavings may inhibit Prototheca growth

• In vitro only
• Bedding samples taken from manufacturing sites and not dairies
• Low sample numbers meant large error bars in results

• 91 farms from 4 regions of Canada. 15 cows from each farm. 3 milk samples were collected from each quarter of the selected cows at intervals of 3 wk for 2 years. Total of 59,167 quarter samples collected
• Farm observations
• Questionnaire on farm management

• CNS IMI incidence of 0.29 new IMI/quarter-month but this was 0.36 when misclassification (due to Se/Sp of culture and other factors) was considered
• Use of sand or wood-based product bedding was associated with lower odds of acquiring a CNS IMI compared with straw bedding

• Milk samples growing >3 types of bacteria were considered contaminated and excluded, meaning > 7,000 samples were excluded
• Mastitis incidence is highly related to management practices and these varied between each farm, so multiple confounding factors were not taken into account

• No difference in lameness incidence or severity by each barn type
• Compost-bedded barns had lower prevalence of Hock lesions (3.8%) than CV and NV barns (31.2% and 23.9%, respectively)
• No significant differences between barn systems for mastitis, culling or mortality

• Small region of USA
• Manual observations are subjective in nature
• 4 visits in 9 months per farm
• Respiration rates only measured in summer
• Much variation in farm management types and sizes

The cleanliness of the cows was evaluated on 3 different body parts: (1) udder, (2) flank, and (3) lower rear legs (1-4 scale), according to the Dutch Udder Health Centre

Hock Lesions (1-4 scale) according to Rutherford K.M.D., Langford F.M., Jack M.C., Sherwood L., Lawrence A.B., Haskell M.J., (2008), Hock injury prevalence and associated risk factors on organic and nonorganic dairy farms in the UK, Journal of Dairy Science, 91, 2265-2274

Bacteriological counts of the bedding materials

Bacterial count in milk

• The percentage of cows having damaged hocks was lower on farms using box compost compared with farms using sand and foam mattresses
• No differences were found between the different types of bedding materials in all the cleanliness variables
• The lying bout tended to be longer when using horse manure vs foam mattresses. The first lying bout tended to be longer on sand vs mattresses or box compost
• Less colony-forming units on sand vs box compost
• Manual observations are subjective in nature

• Only assessed in one season (Spring)
• Each farm visited once for 1 day

New sand, in-use sand, washed sand, recycled sand

• Median numbers of CFU in all types of bedding declined with time, suggesting that suppressive constituents of sand are not inhibited by processing during recycling
• Heating to 80degC reduced counts of coli vs in-use and fresh sand
• Ambient temperature affected non-heat treated sand contamination
• The heating process allows heat-sensitive microorganisms to suppress the E. coli

• Only from one farm
• Sand only and no comparisons with other types of bedding
• Only cultured and studied E. coli O157:H7

the majority used a cross-sectional study design (n = 28) or a combination of cross-sectional and before-and-after designs (n = 6). One study used strictly a before-and-after study design  and only one study used a longitudinal cohort design

• Five databases (PubMed, Medline, CAB, Agricola, and Web of Science) were searched on April 22, 2009
• Forest plots
• Review

1. Intervention studied was a management practice applied or observed at the herd level and used as an udder health control strategy;
2. SCC was measured using cell counting methods rather than California Mastitis Test (CMT) or Rapid Mastitis Test
3. Mean 305-d milk production of the herds studied was ≥7,000 kg; and
4. Mean herd size of the herds studied was ≥40 milking cows

• Freestalls had lower SCCs than pasture and bedded pack
• Sand > mattresses > decreased bedding moisture on their impact on decreasing SCC
• Manure packed systems had lower SCC than other types (from one paper)

• Focused on many other management practices and impact on SCC rather than focusing on bedding type
• Over time, technologies have allowed dairies to have better automated systems and recycling of bedding etc. so when reviewing articles, this has been a challenge to choose what has more weighting.

Continuous video surveillance collected the following data:

Time lying in the stall, time standing in the stall, time standing in the alleys (including drinking), time feeding, and time milking (time out of the pen for milking and transit) during a 24-h period were measured for each cow

relationships of components of the time budget with herd-level and cow-level fixed effects

• Lameness affected time feeding, with the greatest feeding times observed in non-lame cows
• As lameness severity increased, time standing in the alley decreased
• Time standing in the stalls was greater in the mattress herds than sand herds
• Cows in MAT herds spent significantly less time lying, with shorter lying bouts than did cows in SAND herds

• Not randomised,
• Small sample size
• Not including summer (hot) data
• Manual observations are subjective in nature

average of 96 cows per farm and a total of 25,358 cows

From Mar 5 to Sept 5, 2007

• Rubber mats or mattresses as the stall base had a higher percentage of cows with severe hock lesions vs dirt base
• Compared with farms with dirt base stalls, farms with rubber mats and mattresses had more cows standing fully in the stall
• 50% of cows were lying down on farms using sand bedding compared with 40% on farms using straw, sawdust, or dry or composted manure

• Only herds on a national management program were selected, creating potential bias in the results
• Questionnaire allows for subjective assessment
• >140 manual assessors, so high risk of subjective observations and opinions

Sand from one dairy heat and chemical treated to see whether or not Mycoplasma spp. survived

• Mycoplasma was identified from bedding sand, dirt and recycled sand in 3 of the 9 farms and most samples from those farms
• Mycoplasma appeared to survive in deep bedding samples for up to 8 months, peaking in cycles relating negatively to temperature and positively to moisture (deep samples were always moist)
• Treatment with higher concentrations of disinfectants, 2% chlorhexidine and 0.5% sodium hypochlorite, eliminated 100% of Mycoplasma from contaminated bedding sand

• Only one pathogen studied
• Farms mostly used sand with only a few on straw
• Different pen systems investigated (stalls, pack and dry lots)

concrete, sand (particle diameter 0.1 to 0.6 mm), and soft rubber mats

Each group of cows was allowed to choose between 2 of the 3 different types of materials, rotating around all material types. They were allowed 10 days for preference testing after a 5 day enforcement period so they were used to all materials.

• Cows were observed lying down more often on rubber mats than in concrete stalls
• No overall preferences observed between rubber mats and sand stalls or between sand and concrete-based stalls

Time-lapse video footage

Lying times and number of lying bouts

Time budgets for other activities

• cows on mattress stalls had significantly lower lying bout durations than cows on pack mat stalls
• Stall standing time increased with increased lameness on all bed types
• Stall standing for non-lame cows was longer in mattress stalls than pack mat (sand)

• Small sample size for 3 groups of cows
• 4 farms mean management differences may affect results
• Hard for observers to define slightly and moderately lame cows as separate groups

• Lying times and frequency of lying bouts on sand were longer and more frequent than straw
• There were no differences in the dirtiness of stalls between the materials
• Cows housed on straw were dirtier than those on sand
• The severity of hock lesions was lower for cows on sand than for cows on straw though not the incidence (no difference)
• Overall hoof health was greater for cows on sand

• Small sample size
• Leg and hoof injuries inspected only at beginning and end of experiment (1-2 years in between)

Bacterial growth

Nutrients within bedding types

• For pneumoniae, peak growth occurred within 24 hours in all types of bedding and was stationary afterwards
• For growth of pneumoniae, the order was as follows: CS < SH < RS < DS
• For Enterococcus, growth appeared to decrease over the 72 hours in all types, decreasing greatest in shavings
• Growth of faecium was greater in RS and DS than growth for CS or SH
• pH in bedding types: ranging from lowest to highest as SH (4.27) < CS (8.15) < RS (8.37) < DS (8.90)
• High Carbon or pH may be associated with better conditions for bacterial growth

• Only two species of bacteria studied

• Twenty-six percent and 17.8% of operations reported a BTSCC <200,000 cells/mL and >400,000 cells/mL, respectively
• Farms with >500 cows and higher milk production than 9000kg/cow 305d lactation had lower BTSCC values
• Mattress, sand and newspaper bedding were associated with lower BTSCC. Composted manure was associated with higher BTSCC

• Herds in national monitoring scheme so may have included bias to better managed herds
• Questionnaire method increases risk of subjective assessment

6 farm samples used for CS and 6 for RS in Winter

4 samples for RS and 7 for CS in Summer

Samples taken for 6-8 days, twice

Dry matter and organic matter were also assessed

• RS and CS had the same bacterial counts when compared at any sampling time (winter or summer)
• CS had higher DM than RS in winter and summer
• RS had higher organic matter than CS in winter and summer

• Paper not clear whether the same farms were chosen to compare winter and summer samples (suspected that they were not as different numbers of farms were used)
• Huge variations in types of bacteria by farm indicate large variations by facility

12 herds of 150-450 cows were observed but 6 herds using mattresses and 6 herds using sand were chosen for lameness and lying time observations. 602 cows in total were observed with 304 on mattresses and 298 on sand.

Sand or rubber crumb filled mattresses.

• Lameness prevalence in MAT herds was more than 3 times that observed in SAND herds
• No significant difference was found between cow comfort indices and lying times in different systems

• Herds representative of local area but not of USA
• Variations in stocking density were identified across the sample population
• Variations in lameness across herds may have affected lying times

• Over time, the stall surface took on a concave shape, and bedding depth was lowest at the center of the stall and highest at the edges. The loss of bedding was greatest after new bedding was added, and sand levels continued to decline during the entire 10-d period
• For every 1-cm decrease in bedding, cows spent approximately 11 min less lying during each 24-h period
• Duration of lying bouts was shorter when the stalls contained less sand

• Small sample size
• Different observers may show inconsistencies in results

• There were 2 times more coliforms and 6 times more pneumoniae bacteria on teat ends of cows housed on sawdust compared with those housed on sand
• There were 10 times more Streptococcus bacteria on teat ends of cows when housed on sand compared with sawdust
• Udder cleanliness was no different on either bedding type
• There was no clear relationship between teat end counts and bedding counts

• Small sample size

• Fifty percent of the BTSCC samples had a BTSCC <348,000 cells/mL
• BTSCC samples > 200,000cells/ml were 5 times more likely to have a high CNS (coagulase negative staphylococcus) count
• Sand as bedding had significantly lower BTSCC in their BTSCC compared with organic bedding such as shavings, newspaper, and straw

• Farms restricted to NMC milk testing program (therefore may be biased to better management systems)
• Producers that used sand (quite innovative in 2004) may have had better management practices than older-style bedding, indicating that sand itself may not be better than other bedding materials

• There was no significant difference in time lying down in stall
• Time standing in stall was significantly lower in Sand cubicles
• Cows in sand herds had a significantly greater proportion of long lying bouts than did cows in mattresses
• Lame cows stand for longer in stalls than healthy

• Selected herds with high production and potentially better management than overall population
• Manual observations may be subjective

Video footage

Time budgets

• Preference was Sand > Mattress > Waterbed > Concrete
• Cows spent more time lying in Sand but more time occupying rubber mattresses (standing in them)

• Cows had no prior time to get “used” to the bedding types
• At 100% stocking, some cows had no choice on cubicles and lower social hierarchy cows may have been forced to lie on worse bedding
• Some cows may have visited multiple types of bedding so modelling lameness of lesions would have been affected
• A robot milked herd, so position of bed in relation to robot may have confused the data
• Manual observations may be subjective

• Bacterial counts in organic materials (sawdust, straw) were higher than inorganic (sand) materials
• Klebsiella counts were higher in sawdust vs straw
• Streptococcal counts were higher on straw than sawdust
• Gram –ve counts were higher in Summer

• Herds were selected with lower incidence of aureus and Strep. agalactiaea
• Mixed breeds: 8/9 farms had Holsteins, one had Jerseys. There may be breed preferences due to cow size vs stall size.

Bacterial counts from bedding

Klebsiella and Enterobacter, Citrobacter and coliform spp. cultured

• Composting manure (when fresh) effectively reduces the coliform counts to near zero
• When the manure becomes moist and warm, these counts can return to levels found on contaminated bedding.

• Samples only from one farm and unknown over which period
• Only Klebsiella and Enterobacter spp. measured

Cow hygiene (1-5 scoring system) from Reneau J.K., Seykora A.J., Heins B.J., Endres M.I., Farnsworth, R.J., Bey R.F., (2005) Association between hygiene scores and somatic cell scores in dairy cattle, Journal of American Veterinary Medical Association, 227, 1297-1301

• Cows on mattress systems and waterbeds had higher skin lesion scores than sand bedding
• Cows on mattress and waterbed systems had lower hygiene scores than sand bedded systems

• No dairies used recycled manure or digester solids
• Average herd size was high (>800 cows) and milk production high, so management practices may have been biased towards better, more efficient systems
• Subjective, observer based scoring systems

1. bed occupation time of the free stalls per day;
2. average duration of each lying event per day;
3. duration of periods spent standing in the free stall, with 2 or 4 feet per day (respectively);
4. number of interrupted attempts at lying down per hour ;
5. mean duration of a single lying bout in minutes; and
6. time spent ruminating in the free stall, during the lying down time per day.

Cow hygiene

Milk bacterial count

• Cows appeared to occupy the EVA mats more than other types
• Lying times decreased with increasing temperature-humidity index (THI) in EVA mats
• Cow comfort index showed greatest consistency over all THI recordings with wood shavings
• Cows tended to stand more, rather than lie, in mattresses, especially the PVA type

• Very small sample number
• May be some bias from company selling EVA mattresses as results tend to differ significantly from other literature reports (lying times on sand and manure vs. mattresses)

Milk yield and milk bacterial count

• More time was spent lying on sand and straw yard than mattresses
• Time of lying bout was no different between bedding types
• Mattress stalls were used for standing and sand for lying
• Greater milk yield in sand vs mattress bedding
• Manual observations may be subjective

• Small sample population

Herd clinical mastitis, SCC, high SCC prevalence (% of herd

>200,000 cells/mL SCC), and BTSCC), locomotion, hygiene, and hock scores

• Only 50 cows per farm scored
• Observer, subjective scoring

1 = concrete, softness of 0 mm;

2 = rubber, softness of 1 to 8 mm;

3 = soft mats, softness of 9 to 16 mm;

4 = multilayer mats, softness of 17 to 24 mm;

5 = mattresses, softness over 24 mm

• Herds on concrete free-stall bases yielded 6,727 ± 146 kg of milk from 5 to 305 days in milk. In comparison, herds showed a decrease of 0.3% on rubber, an increase of 2.4% on soft mats, an increase of 4.5% on multilayer mats, and an increase of 3.9% on mattresses
• Compared with concrete, the hazard ratio of teat lesions was less on rubber, soft mats, multilayer mats, and mattresses

• Farmer-lead, subjective questionnaire
• Milk yields modelled using Woods equation rather than actual data
• Farms with lower yields may have genetic and breed differences, affecting both production and locomotion

• 4% cows were sound (locomotion score=1), 6.9% were mild/moderately lame (locomotion score=2), and 2.7% were severely lame (locomotion score=3).
• 3%) had no hock lesions (hock score=1), 10.1% had mild lesions (hock score=2), and 2.6% had severe hock lesions (hock score=3)
• 2% were BCS <=2.25 on a 1-5 scale
• Lower locomotion scores (2 or lower) were associated with larger operations, pasture and sand stalls vs straw, compost or matting
• Lower hock scores were associated with dry lot systems

• Multiple assessors using subjective scoring systems but trained to single scoring system

• Greater Lying times and lower hock scores were associated with waterbed mattresses
• Rumination time was greater on rubber mattresses
• Milk yield was not significantly different between the two systems

• Small cow numbers
• Unable to determine whether or not conflicting interests are present

1. Concrete floor (covered feeding area) and brick paved (open)
2. Concrete (covered feeding area) and sand bed (open)
3. Rubber mat (covered feeding area) and sand bed (open)
4. Rubber mat (covered feeding area) and brick paved (open)

Lameness (1-5 scale)

Walking speed

• Walking speed was greater in sand bedded groups
• Lameness score was greater in brick paved groups, with greater severity also

• Very small group numbers
• Conclusions made from underpowered findings
• Subjective scoring scale

Claw lesions score and severity

• No significant differences in locomotion between different types of bedding material, flooring system, breed, visit number, observer and time since last trimming
• Statistically significant differences in the prevalence of claw disorders between compost bedded and freestall cubicle barns were found for white line disease (WLD; 20.4% and 46.6%, respectively), heel horn erosion (HHE; 26.9% and 59.9%, respectively), concave dorsal wall as a result of chronic laminitis (6.5% and 15.9%, respectively) and for interdigital hyperplasia (0.2% and 3.1%, respectively)

• Subjective scoring systems
• Comparing bedding types when flooring systems would differ and barn design may have influenced the study outcomes (whether or not there were sharp turns, steep grades, flooring flaws etc.)

Hair loss on lateral aspect of hock (using hock map), Hock lesion (score) and lameness (score)

BCS, milk yield, duration of housing

• Cows on concrete based cubicles with sand or chopped straw-based bedding had lower hair loss on hock than those on concrete with whole straw/rape straw dressing
• Cows on concrete based cubicles with sand or chopped straw-based bedding had lower hair loss on hock than those on mattresses with whole straw/rape straw dressing

• Purpose of the paper was to compare the hock map versus the scoring system for analyzing hock lesions rather than the bedding material itself

• Dairy data records:
• milk yield,
• pregnancy after first insemination (P/FAI),
• somatic cell counts (SCC),
• incidence of clinical mastitis,
• incidence of metritis and cytological endometritis, and
• mortality/culling rate

• No differences in mortality, fertility parameters or uterine health
• Lower clinical mastitis rates in compost bedding
• Cows yielded avg. 760kg more milk in the first 100DIM from the compost bedded system

• Higher milk yield could be double-counted from the lack of mastitis (therefore less dumped/future milk losses)
• Milk yield error limits crosses over between the two bedding types, making the result less significant
• Low numbers to get sufficient power for good conclusions on yield and fertility parameters (altering things in the dry period are a long way in time from getting cows pregnant and many other factors can influence the outcomes, so making conclusions would be extremely difficult)

• Hours of lying were longer on compost than freestalls
• Sound cows and lame cows lay for the same times on compost but sound cows lay longer than lame cows in freestalls.

• Very small numbers
• Many environmental changes if going from freestall + pasture to compost, loose-housing. Feeding areas, water access, design of barn, heat abatement could all affect outcomes
• Cows were on the freestall system for 495 days and compost for 132 days

Dairy software records

• Severe lameness increased with the percentage of stalls with fecal contamination and with use of sawdust bedding, and decreased with deep bedding, sand bedding, herd size, and rearing of replacement heifers on site as well as frequency of manure removal in the pen per day
• Deep bedding and sand were highly correlated as most systems used both on the same site.

• Subjective scoring by multiple assessors
• Modelled data from records

Hygiene (cleanliness score)

Lying times

• Lying times and number lying down were longer on matting than manure straw bedding and sand

• Very small cow numbers
• No information on heat abatement, as this could have affected lying times (thin, rubber matting may be cooler than manure or sand). Turkey has a climatically high THI.

• There were no significant differences between mean of locomotion scores on concrete, sand or rubber bedding in this study

• Subjective scoring system
• Varying herds with very variable environments and management systems
• Comparing bedding types with more than one barn type (freestall and tie-stall) means that it is difficult to make conclusions with so much variation

Bacterial contamination of bedding

Lameness, BCS, hock lesions and hygiene

• Bedding pack material averaged 15.8% for total C, 0.93% for total N, 17.8 for C:N ratio, 37.3% for dry matter, 8.83 for pH, 4.25 mg/kg for nitrate, 955 mg/kg for ammonium, 15 g/kg for total potassium, 2.8 g/kg for total phosphorus, 8.5 dS/m for EC, 31.7 degrees C for pack temperature, 7.6 degrees C for outside temperature, and 9.42 million CFU/mL for total bacterial counts in the bedding
• All types of substrate were considered suitable but regular tilling and replacing were recommended

• No information on cow factors

California milk test (CMT)

• Decreased odds ratios for subclinical mastitis were found on farms using systems other than rubber mat/concrete beds

• Authors admitted that many of the associated factors and odds ratios were likely to be related to system management variation between farms rather than the factors in the paper themselves.

Records from dairy software

Lameness score (1-5 scale)

• The mean prevalence of clinical lameness was 24.6%
• Prevalence of lameness in parity one averaged 12.8% and increased by 8 percentage points each parity.
• Lameness was lower in freestall herds with sand stalls (17.1%) than in freestall herds with mattress stall surfaces (27.9%)

• Lameness scored by multiple assessors and a subjective system

• Of the 12 cows used in Experiment 1, 10 preferred sawdust before and nine after the restriction phase. During the restriction phase, average lying times and number of lying events during the restriction phase were significantly lower for the sand-bedded stalls (P<=0.05), and standing times were higher on mattresses (P<=0.05), compared with sawdust
• After restriction, half of cows preferred sawdust and half sand.

• All cows had been housed on sawdust for their previous lactation (although they were given experience of all types of bedding before the experiment)
• Very small population number

Experiment 2: 2cm deep rough or fine sand

• Lying times were shorter and fewer lying bouts were observed on sand than on either straw or rubber matting in Experiment 1
• In winter, cows preferred to lie on straw rather than rubber matting
• When forced to lie on sand, cows had no preference to either rough or fine sand or lying in the concrete alleyway.

• Before the experiment, all cows were housed on rubber matting covered with peat and straw, so some may have had previous preferences
• Very small population numbers
• Stocking density was different in winter (100%) to summer (50%)
• Sand layer was 20cm deep which seems low compared with other papers reviewed in this knowledge summary
• Manual observations may be subjective

• 73% of cows had at least one hock lesion
• 91% cows had lesions on mattresses
• 24% cows had lesion on sand

• Subjective scoring methods
• Only 4 farms out of 20 had sand, 10 on mattresses and 6 on sawdust, with 185/1752 cows on sand.

• Cows on matts had longer lying times with more frequent bouts of lying
• Cows on matts had lower standing times and less time spent eating than soil based bedded cows
• Manual observations may be subjective

• Very low sample numbers

Growth of coliforms on days 0, 2, 5, 8 and 11 on dried wood sawdust, composted washed manure solids (CS), chopped ryegrass straw, and shredded paper covering gypsum wall board in freestalls under farm conditions, then:

Treating recycled manure solids with hydrated lime, copper, zinc and germicide to control coliform growth

• Coliform growth on all types of bedding materials reached faecal output levels within 2 days.
• Treatment of bedding with any of the applied materials did not affect bacterial growth in the bedding material

Sand, 12-mesh limestone and treated with pine disinfectant, oat straw and cedar sawdust. Mulched newspaper

• Sand, limestone (including treated) both had significantly higher growth after 6-54 hours incubation than organic bedding (straw, sawdust, newspaper).
• Organic materials (sawdust, newspaper and straw) showed bimodal growth patterns, declining at 30-54 hours and increasing again beyond inorganic levels after 96-120 hours

• Unknown which bacteria grown and what season, temperature, humidity and levels of contamination beforehand.
• From Table 3, it is inferred that there were 10 samples taken from unused pens and 30 from used, totaling 40 samples but this was for E. coli specifically. The paper is not clear on methodology how many samples were taken for all bacterial species.