1:3: 6.: 5.3. .15.; . . . 1 .. ll . 1 - .- J. g...ffi§§ég§fi r. .113..5.‘31w..m....um.wr....mwws333.3%.... I ...- .. .. .- nfinrfv: (1.? .. : .._.&..%.13.gs&an§h. 3%.}...Jma.qu3...«$.mmmfimm..+ummuwm...w..3...v.m..u.mm.w.......mm...x.m.ou... .uflfinbt Wm”... . I. .. .. . ”3833.33.3wmez4 u. _. 33%. .3 .1. . 1.333.333.3331? . ...... . 34.393.25.31 .H . 3g . . . Iww . m. 1...... ........ ,. . a..n............-.: H . . . . .. .. . ., . 2.. .3... S St .u. . y . 4 _ I?! lv‘u . tan... . .... .WWQunu... Na .. . ....n.........h.1....r. 3.113.331 11$ MN. I. ... .YI. n I .c 91-"? .Ynfihi: 3.130 .‘thAwfinnofikfva 0013' I a. 1 :QOLTA .50.! 10A . n . I o .4 V iv {In oumll; . In" . I l. . . . . .- 34:33.33..- 3.. 3m 2...... . a. 3...“ 159.15.. \nn? 0. .7' 3:31 o . - . rzm x..- &§. 1 kg/day) used has been shown to reduce incidence of clinical mastitis and facilitate maintenance of clean dry stalls (Saloniemi, 1980). However Rendos et al. (1975) had shown straw contained a large population of streptococci and staphylococci per gram of straw. Use of recycled manure solids as bedding material is of special interest to western states because of expense of traditional (sawdust, straw, wood chips, etc.) bedding materials. Carroll and Jasper (1980) reported dried or composted-dried manure was an adequate bedding material if it was kept dry and not damp with urine and feces. However good composting temperatures were only reached in the interior of the manure pile (Carroll and Jasper, 1980). Allen et al. (1980) conducted two separate studies on the practicality of recycled manure solid bedding. During the first studies, freestall bedding was changed from sawdust to recycled manure solids at 20% dry matter. Over the next year, CMT scores increased, incidence of clinical mastitis increased and coliform.mastitis increased from 7 to 46 percent of total cases. However, occurring simultaneously with bedding change was a change in milking equipment, milking hygiene and dry cow treatment. Due to extent of mammary health problems, cows were rebedded on sawdust. Two years later Allen et al. (1980) conducted a more controlled study using recycled manure solid as freestall bedding. In this study, manure solids were composted for six weeks 43 prior to use. This procedure increased dry matter from 20 to 60 per- cent and eliminated all coliform organisms from outer 48 inches of compost pile. Allen et al. (1980) found no significant differences between sawdust and recycled manure solid bedding. Effect of milking system The milking machine comes in direct contact with the cow's teats two or more times daily. Improper functioning or use of the milking system has been suggested as preconditioning the cow to teat injury, to increased incidence and spread of mastitis and to higher somatic cell counts. There are three general types of milking systems; bucket, barn pipeline and parlor pipeline. Saloniemi (1980) found incidence of mastitis was significantly (p §_.05) higher with bucket systems than pipeline systems. Downey et al. (1977) found the average somatic cell count for bucket systems to be highest (393,000 r 17,700 cells/ml) and parlor systems in general to be lowest (359,000 1 23,500 cells/ml). Barn pipeline milking systems which were the most numerous type of system (76.4% of total) were associated with intermediate cell counts (385,000 r 8,300 cells/ml). Schultz (1977) also found that parlor milking systems were associated with lower cell counts than other milking systems. was milked in a low line (pipeline located below udder) milking parlor had the lowest cell count compared to other parlors. Cows milked with barn pipeline systems were found to have highest somatic cell counts. Schultz believed the higher somatic cell counts for around the barn pipeline and high line parlor systems were due to the required elevation of milk in these systems. 44 McDonald (1971) had shown high milk pipelines (greater than five feet above udder) resulted in excessive vacuum fluctuations at teat end. High pipelines also require a higher preset vacuum level to maintain good milk flow rate. The higher vacuum level could increase probabil- ity of teat and streak canal injury at end of milking and, thus, increase probability of infection. Downey et al. (1978) reported milking systems less than five years old had lower somatic cell counts than systems of greater age. Saloniemi (1980) found a significant (p.: .05) correlation of 0.41 between milking machine age and incidence of mastitis. He calculated an increase of 0.012 cases/cow/year for each yearly increase in milking machine age. Expansion of a milking system seems to increase somatic cell count. Downey et a1. (1977) studied 26 progressive Canadian dairy herds. They found average somatic cell counts of expanded milking systems to be 394,000 r 14,500 cells/ml and 381,000 r 8,300 cells/ml for systems still operating at their original size. In a larger study, it was found expanded systems had a higher mean somatic cell count (530,000 cells/ml) than systems still operating at their original size (430,000 cells/m1) (Downey et al., 1978). An increasing number of milking systems are being automated. Philpot (1973) reported CMT and microbe content of milk decreased and physical condition of teat ends improved with addition of Surge QTO® automatic take-off devices. By preventing overmilking, milking after milk flow is less than i pounds milk per minute, teat injuries and incidence of new infection are reduced (Mochrie et al., 1953a; 45 Mochrie et al., 1953b; Petersen, 1964; McDonald, 1971; Natzke et al., 1976). However Natzke et a1. (1978) found no significant (p > .05) increase in new infections with overmilking. Various milking system conditions influence teat injury and incidence of mastitis, including status of vacuum pump, vacuum level vacuum controller, vacuum line, pulsator, claw and inflation (liner). Saloniemi (1980) reported a highly significant (p < .001) increase in teat injury and incidence of mastitis as an increasing number of these conditions varied from manufacturer's specifications. vacuum level and vacuum stability have been associated with incidence of mastitis. As vacuum level increases the amount of teat damage increases resulting in an increased infection rate and elevated somatic cell count (Afifi, 1968c; McDonald, 1975; Nicolai et al., 1977; Noorlander, 1977; Galton and Mahle, 1980; Saloniemi, 1980). Static vacuum between 25.4 and 33.0 cm Hg (10-13 inches Hg) is best for minimizing teat trauma during milking (McDonald, l971; Galton and Mahle, 1980). McDonald (1971, 1975) reported increased injury to both teat end and streak canal and teat congestion when vacuum at teat and exceeded 33 cm Hg (13 in Hg). Nicolai et al. (1977) reported no significant (p - .01) differences between cows milked at 10 and 12.5 inches Hg vacuum. However cows milked at 15 inches Hg vacuum had significantly poorer teat condition and higher CMT and DMSCC scores. Afifi (1968c) reported a significant (p < .01) increase in somatic cell count when vacuum levels exceeded 40 cm Hg (15.75 in Hg). Highest somatic cell counts occurred when vacuum levels were 55 cm Hg (21.65 in Hg). Saloniemi (1980) also reported a significant (p < .05) 46 increase in incidence of clinical mastitis when vacuum levels were greater than 50.7 kPa (14.91 inches Hg). However Mochrie et al. (1953a) and Mbchrie et al. (1953b) reported no significant effect of vacuum levels of 10, 13.5 and 17 inches Hg on leukocyte count, presence of mastitis organisms or milk production. The effects of high vacuum level are more pronounced when cows are simultaneously overmilked (Afifi, 1968c; McDonald, 1975). Vacuum levels below recommended levels have also been shown to significantly (p j .001) increase teat injury (Saloniemi, 1980). Any interference with free air flow in a milking system between the teat end and vacuum pump can cause vacuum fluctuation at the teat end during milking. Thiel et al. (1973) found vacuum fluctuation either at teat end or between shell and inflation (liner) of teat cups or large cyclic fluctuations did not cause a significant (p > .05) increase in infection, but vacuum fluctuation at teat end or between shell and inflation combined with large cyclic fluctuations significantly increased infections (p = .01 with constant pulsation rate, p = .001 with fluctuating pulsation rate). Pressure changes within the milking system can cause bacterial laden milk droplets to be introduced into the teat during milking and thus increase risk of mastitis (Thompson, 1980; Noorlander, 1977; O'Callaghan and O'Shea, 1979). To prevent teat or streak canal injury and related increased udder infection, vacuum levels should not fluctuate more than 7.6 cm H (3 in. Hg) at teat end during maximum milk flow when the system is fully loaded (McDonald, 1971). 47 The vacuum controller (regulator) is closely associated with vacuum level and fluctuation. The function of the vacuum controller is to maintain preset vacuum level at a designated location in the milking system throughout almost the entire range of the vacuum pump's capacity to remove air. Smith and Fairbank (1975) compared spring loaded diaphragm, weighted sleeve value and weighted lever controllers. The latter two controllers were found to be adequate if maximum load changes were of low magnitude (i 10 cfm or less). Spring loaded diaphragm controllers were found to be more sensitive and able to handle load changes of large magnitude (3 20 cfm or more). However, increasing pump capacity does not compensate for a poor controller, nor does a more sensitive controller reduce pump require- ment (Smith and Fairbank, 1975). Poor vacuum controller performance significantly (P.: .05) increases clinical mastitis (Saloniemi, 1980). . Vacuum reserve capacity is recommended to be at least 50% above actual measured air requirements (McDonald, 1971). Inadequate effective vacuum reserve leads to irregular vacuum fluctuations and increased mammary infection (McDonald, 1975). Klastrup (1969) reported infection level decreased from 49 to 16 percent as vacuum reserve per unit increased from 0 to greater than 5.4 CFM free air. However Maatje and Rossing (1971) and Saloniemi (1980) found no relationship between cell count or incidence of clinical mastitis and vacuum reserve capacity. Pulsation rate and ratio also effect somatic cell count and incidence of mastitis. Milne (1977) reported milking systems with operational faults, but with properly functioning pulsators, had 48 bulk tank milk cell counts 14% higher than completely efficient milking systems. Milking systems with faulty pulsators had 30% higher bulk tank milk cell counts than completely efficient milking machines. Afifi (1967, 1968c) showed pulsation rates greater than 50 cycles/minute or less than 44 cycles/minute significantly (p < .05) increased somatic cell count. McDonald (1971, 1975) reported pulsation rates of 40-60 cycles/minute were adequate with 60 cycles/minute being the most economical. However Saloniemi (1980) found pulsation rate had no effect on incidence of mastitis. Galton and Mahle (1980) examined the interaction of pulsation ratio and vacuum level. Pulsation ratio of 60:40 (60% milking phase to 40% massage phase) was associated with lowest somatic cell count across all vacuum levels (10 inches (25.4 cm) Hg, 12.5 inches (31.75 cm) Hg and 15 inches (38.1 cm) Hg). A 50:50 ratio resulted in increased incidence of teat injury but not infection rate. Pulsation ratios of 70:30 (70% milking phase to 30% massage phase) or greater increased teat injury and congestion, infection rate and somatic cell count (McDonald, 1971; McDonald, 1975; Galton and Mahle, 1980). These conditions are more severe at vacuum levels greater than 13 inches (33 cm) Hg. Galton and Mahle (1980) hypothe- sized the detrimental effects of 70:30 pulsation ratio were due to inadequate length of massage phase causing incomplete filling of the test cistern, undue stress on teat tissue and forcing of pathogens through the streak canal. Hoare et al. (1979) reported somatic cell count significantly (p < .05) increased as proportion of massage phase of cycle decreased and milking phase increased. Somatic cell count 49 increase was attributed to increased injury and spread of infection during long milking phase. Saloniemi (1980) found as milking phase increased there was a significant (p §_.05) increase in teat injury. Recent reports have implicated inflation (liner) design, material of construction, speed of closure and opening, and admission of air between inflation and teat (slip) as major causes of intra- mammary infections. Improper inflation design influences teat damage and erosion, teat massage, flooding of milk, milking efficiency and contamination of teat orifice (McDonald, 1971; Noorlander, 1977). There are two basic inflation types, narrow bore and wide bore. Narrow bore inflations (3/4 inch or less in diameter) have been shown to significantly (p < .01) reduce udder irritation, lower bacterial infection, significantly (p < .05) reduce clinical mastitis and amount of machine stripping and milking time when compared to wide bore inflations (Dillion et al., 1969; McDonald, 1971). However inflation size must match shell size; i.e. wide bore inflations used with wide shells, narrow bore inflations used with narrow shells. Scanning electron microscopy has revealed the material from which the inflation is constructed is important. At 1,000 milkings the inner surface of rubber inflations contain deep cracks and caverns. By 5,000 milkings bacteria were common in these openings. However Silastic® inflations remained smooth after 1,000 milkings and only began to show signs of wear, but no bacterial contamination, at 5,000 milkings (Heckman and Noorlander, 1980; Nborlander and Heckman, 1980). 50 The speed of inflation closure and opening has been shown to be partly responsible for impinging of milk backward against the teat (Noorlander, 1977). Rapid closing and opening may also accentuate large cyclic vacuum fluctuations increasing incidence of infection (Thiel et al., 1973). Admission of air between inflation and teat (liner slip) is associated with 50—60% of new infections (O'Shea et al., 1979). O'Shea et al. (1979) compared Gascoigne single stretch inflations with Alfa-Laval Liner 960000-1. They reported significantly (p < .05) more infection with Gascoigne single stretch. They concluded the higher infection rate with those inflations was due to greater liner slippage causing milk droplet impact on teat ends. Westgarth (1977) had been able to reduce new infections by 96% by using shields within short milk tubes during a short duration trial. However, new infections were only reduced 10% during a six month field trial. O'Callaghan and O'Shea (1979) and O'Shea et al. (1979) reported impacts arising from slip were not prevented by shields, large claws or changes in pulsation characteristics. Thompson (1980) also found no significant effect of claw size on impacts. However venting of claw decreased impacts due to improved emptying of claw and constant out flow toward milk line (Thompson, 1980). Effect of milking hygiene practices Milking hygiene practices have been associated with both new infection rate and somatic cell count. Early reports advocated full milking hygiene which included washing the udder with either separate 51 paper towels or separate boiled cloth plus disinfectant solution, milkers wearing rubber gloves, pasteurization of teat cup clusters between cows and teat dipping immediately after milking. Full milking hygiene has been shown to reduce new infection rate 58%, infections due to Staphylococcus aureus 62% and infections due to streptococci 70% compared to limited hygiene (washing udders with common cloth using only water) (Neave et al., 1969). Previously Neave and Oliver (1962) reported washing udders with a common cloth significantly increased the chance of recovering organisms from teat skin. Schultz (1977) found use of individual paper towels during udder washing lowered somatic cell count. However Moxley et al. (1978) reported use of individual paper towels had no effect on somatic cell count. washing the udder with soap or sanitizer and water has been shown to significantly (p < .05) lower bulk milk somatic cell count (Hoare et al., 1979), but later work showed pre-milking washing with either water or water with 2% chlorhexidine solution did not signif- icantly (p > .05) reduced new mammary infections (Sheldrake and Hoare, 1980a; Sheldrake and Hoare, 1980b). Drying of teats and udders after washing has been recommended, but the effect of this practice has not been clearly determined. Moxley et a1. (1978) reported udder drying with separate paper towel had the second greatest effect on lowering cell counts (-44,000 cells/ ml). Hoare et al. (1979) found udder drying increased somatic cell count. However in this study drying was done with a cloth or sponge in a weak disinfectant, not individual paper towels. 52 Neave et a1. (1969) and Moxley et al. (1978) reported pasteurization or disinfection of teat cup clusters between cows did not affect new infection rate or somatic cell count. Bushnell et al. (1978) reported addition of backflushing in a commercial dairy gave a rapid decrease in clinical mastitis and eliminated mycoplasma mastitis, but over a two year period subclinical mastitis increased from 20% to 40% due in part to failure of the backflush system to maintain 25 ppm iodine concentration in the sanitation cycle. Machine stripping has not been shown to significantly increase somatic cell counts (Goff and Schmidt, 1967; Afifi, 1968c). Teat dipping has been shown to have the greatest effect on reducing new infection rate and somatic cell count. Neave et al. (1969) reported teat dipping resulted in a seven-fold decrease in new streptococcus infections during a nine week period. Langlois and Pyles (1975) reported use of the commercial teat dip Bovadine® and Chlorox Liquid Bleach® (4% chlorine) showed Chlorox® treated group had 8% fewer clinical mastitis cases and fewer Staphylococcus aureus infections (Grant et al., 1976). Teat dipping with 5,000 mg available iodine/liter has been shown to significantly (p < .05) reduce new mammary infections and reduce both Staphylococcus aureus pOpulation on teat ends and new infections but was not effective against Streptococcus dysgalactiae (Sheldrake and Hoare, 1980a; Sheldrake and Hoare, 1980b). Moxley et a1. (1978) showed teat dipping significantly (p = .01) lowered somatic cell count by 53 70,300 cells/ml. Hoare et al. (1979) also showed teat dipping significantly (p < .05) lowered bulk milk somatic cell count. The bovine udder is most susceptible to infection at the beginning of the dry period mainly because bacteria are able to penetrate the streak canal more easily (Cousins et al., 1980). Dry treating, infusion of oil—based antibiotics into mammary gland at cessation of daily milking, has been shown to reduce new infectnmi rate and cure existing infections. Oliver et a1. (1962) reported dry treating resulted in very nearly complete protection against staphylococcus and streptococcus infections. Rindsig et a1. (1978) reported a 3.1% new infection rate for complete dry treatment and 6.5% for selective dry treatment program. This compares to 10 to 15% new infection rate without dry therapy (Natzke, 1971). Dry treatment has been shown to eliminate 85.4% to 100% of existing infections depending on type of drug, type of infectious organism and whether therapy is complete or selective (Meaney and Nash, 1977; Rindsig et al., 1978). Use of a lactating cow product for dry treatment is not as beneficial. In comparing lactating and dry cow products used at drying-off, Philpot (1973) found dry cow products had a 2.24 times greater cure rate against staphylococcus and 1.09 times greater against streptococcus than lactating cow products. Use of teat dipping ang_dry cow treatment has been shown to be more beneficial than either practice alone. Cows which were both teat dipped and dry treated had 20.2% fewer Streptococcus agalactiae infected quarters, 6.6% fewer Staphylococcus aureus infected quarters 54 and a 1.03 new infections per cow per year compared to 1.50 for the control group which were dry treated only (Eberhart and Buckalew, 1972). The former group also had lower WMT scores after freshening. Schultz (1977) reported herds that dry treated only had higher somatic cell counts than any other combination of dry treatment and teat dipping. Schultz (1977) concluded that dairy farmers were attempting to control mastitis with dry treatment alone, without other good management practices. Hoare et al. (1979) reported bulk milk somatic cell counts were significantly (p < .05) lower for farms teat dipping and dry treating all cows than farms using any other combination of teat dipping and dry cow treatment. MATERIALS AND METHODS A series of models were constructed to determine effect of various bovine age, parity, stage of lactation, milk production, and milk fat production (measured in percent), environmental, and managerial factors on somatic cell counts (SCC). The first model constructed determined the effect of various bovine and seasonal factors on somatic cell count (SCC). Cow age, parity, stage of lactation, and test day milk production, percent fat, and somatic cell count were obtained in cooperation with Michigan Dairy Herd Improvement Association (DHIA) for all cows in Michigan enrolled in the somatic cell count option from November, 1978 through January, 1981. SCC was determined using a Foss-O-Matic® manufactured by A/S N. Foss Electric, Denmark. The model, based on this information, was: Y = u + Herd + Cow + Milk + Lac + Dim + Mo + Fat + Age + Hsize + e where: Y is natural log of reported somatic cell count u is population mean Herd is hard effect Cow is cow effect Milk is milk production effect Lac is parity effect 55 S6 Dim is stage of lactation (days in milk) effect Mo is seasonal (month of sample) effect Pat is percent fat effect Age is cow age effect Hsize is size of herd effect e is random error effect All factors except error were fixed. The second series of models were constructed to determine the effects of various managerial practices and farm conditions on somatic cell counts (SCC). It was determined that the most cost efficient method of obtaining information concerning managerial practices and farm conditions would be by a series of survey question- naires. A series of survey questionnaires were developed concerning milking procedures and practices, milking eqUipment, milking herd management, dry cow practices, and calving practices (Appendix A). The questionnaires were checked for clarity, accuracy, lack of bias in either questions and/or responses and ability to code and analyze data obtained from the survey. Necessary modifications were made prior to distribution of any of the questionnaires. Due to increased postal rates and budgetary constraints, it was not possible to mail survey questionnaires and a preaddressed stamped return envelope directly to dairy farmers enrolled in the SCC Option. After consulta- tion with representatives of parties involved, it was decided to have DHIA technicians distribute the questionnaires to dairy farmers enrolled in the SCC option on test day. Completed questionnaires were picked-up and returned either that day or during the subsequent 57 test period. Individual technicians were given an information packet explaining the research program, a copy of each questionnaire, and a copy of an eXplanatory letter to be mailed to dairy farmers enrolled in SCC Option (Appendix B). This information packet was given to them during a meeting held during March, 1980. Prior to distribution of survey questionnaires by DHIA technicians, dairy farmers enrolled in the SCC option were directly mailed a one-page letter briefly explaining the questionnaires and requesting the dairy farmer to complete and return questionnaires as soon as possible (Appendix C). The first set of questionnaires also contained a release form to permit use of the dairy farmer's DHIA records (Appendix D). Question- naires were distributed and returned between April and June, 1980. Questionnaire response rates were milking practices 43.7%, milking systems 43.4%, milking herd management policy 35.2%, dry cow and calving practices 35.1%, and housing 35.0%. Despite the decline in response rate, it is still much higher than accepted range of 10-15% for questionnaires of this type. As questionnaires were returned, they were coded onto computer sheets and later keyed into the main computer. Prior to analysis, it was necessary to transcribe the data because of computer coding differences between DHIA's Honeywell system and Michigan State University's Control Data Corporation's 1600. Due to a DHIA computer operator error, addition of protein to reports, and a change in DHIA report forms, DHIA data was found to be written in three distinct formats. To facilitate analysis, a single format was developed and data reformated. 58 Actual recorded somatic cell counts, in 100,000 cells/ml, plus one were transformed into natural logarithms (Loge). This transformation was used because of its proven linearity, equality of mean and median, normal distribution, uniform variance, and mean in midscale characteristics. Detailed discussions of somatic cell count transformations are found in Ali and Shook (1980) and Shook (1982). Analysis of all models was done by Statistical Analysis System (SAS) GLM (General Linear Models) with absorption of herd and cow within herd effects. Absorption removed effect of these factors from the analysis. Since all farm condition and management practice models were based on results of a single survey conducted over a three month period, it was decided to use mean somatic cell counts, mean daily milk production, and mean days in milk during analysis based on these models. All models were checked for normality of errors and variance. Intrafactor comparisons within each model were done using Bonferroni-t test. Due to the broad scope of information gathered, it was not possible to construct a single model including all farm conditions and managerial practices of interest to determine their effect on somatic cell count. Thus a series of models were used. The first series of models looked at effect of farm conditions on somatic cell count. Farm conditions were divided into two categories; milking system and housing system. In the milking system models, the symbols were defined as follows: Y 8 estimated effect on log of mean somatic cell count Mlk = mean daily milk production effect 59 Mdim = mean days in milk effect Lac - parity effect Sys - milking system effect Equip - milking system equipment brand effect Lineht - milk pipeline height effect Vaclev - line vacuum level effect Vaccon - brand of vacuum controller effect Ldsgn - vacuum line design effect Lmat - material of vacuum line construction effect Unit - number of milking units effect Pipedi a milk pipeline diameter effect CFM 8 CFM per milking unit effect Pultyp a pulsation type effect Pulrat - pulsation ratio effect Infla - inflation brand effect Auto 8 type of milking system automation effect e 8 random error effect The milking system models were: Y = Mlk + Mdim + Lac + Sys + Equip + e Y a Mlk + Mdim + Lac + Sys + Lineht + Sys(Lineht) + e Y = Mlk + Mdim + Lac + Sys + Lineht + Vaclev + Lineht(Vaclev) + e Y = Mlk + Mdim + Lac + Sys + Lineht + vaccon + Sys(Lineht) + Sys(Vaccon) + Lineht(Vaccon) + e Y a Mlk + Mdim + Lac + Sys + Ldsgn + Lmat + Ldsgn(Lmat) + e Y = Mlk + Mdim + Lac + Sys + Unit + Pipedi + Unit(Pipedi) + e In each fixed. Housing Y: Y: Mlk Mlk Mlk Mlk + Mdim + Mdim + Mdim + Mdim + + + of these milking 60 + Sys + CPM + e Lac + Sys + Pultyp + Pulrat + Pultyp(Pu1rat) Lac + Sys + Pulrat + Vaclev + Pulrat(Vaclev) Lac + Sys + Infla + e Lac + Sys + Auto + e system models all factors except error were In housing system models, the symbols were defined as follows: Y a estimated effect on log of mean somatic cell count Mlk Mdim Lac House Bed Mat Matbed Wcalf Scalf e mean daily milk production effect mean days in milk effect parity effect housing system effect bedding material effect maternity facility effect maternity facility bedding effect fall to spring calving conditions effect spring to fall calving conditions effect random error effect system models were: Y I Mlk + Mdim + Lac + House + e Y - Mlk + Mdim + Lac + House + Bed + House(Bed) + e Y = Mlk + Mdim + Lac + House + Mat + Matbed + Mat(Matbed) + e 61 Y = Mlk + Mdim + Lac + House + Wcalf + Scalf + Mat + Matbed + Wcalf(Mat + Matbed) + Scalf(Mat + Matbed) + e In each of these housing system models all factors except error were fixed. Management practices were divided into five categories; milking hygiene practices, treatment of mastitic cows, dry cow treatment policy, free stall maintenance, and purchase of replacement animals. Symbols used in the management practices models were defined as follows: Y3 Mlk = House 2 Prep = Lag = Wash = Dry = Dip 8 Spray = Bdip 8 Rinse - Masmlk = Lactrt 8 Trtad = estimated log of mean somatic cell count mean daily milk production effect mean days in milk effect parity effect milking system effect housing system effect prep time effect prep-lag time effect udder washing method effect udder drying effect teat dipping effect teat spraying effect brand name of teat dip effect teat cup liner rinsing between cows effect when mastitic cows milked effect type of treatment for mastitic lactating cows effect site of treatment administration effect 62 Numtrt - percent of total herd dry treated effect Btrt a brand dry treatment product used effect Bed - bedding effect Bedadd I frequency free stall raking effect Bedclean - frequency free stall cleaning effect Ovsc a open vs. closed herd policy effect e - random error effect Management practices models were: Y Y Mlk + Mdim + Lac + Sys + Prep + Lag + Prep(Lag) + e Mlk + Mdim + Lac + Sys + Wash + Dry + Dip + Spray + wash(Dry) + wash(Dip) + Wash(Spray) + Dip(Wash + Dry) + Spray(Wash + Dry) + e Mlk + Mdim + Lac + Sys + Wash + Dry + Bdip + wash(Dry) + wash(Bdip) + Bdip(Wash + Dry) + e Mlk + Mdim + Lac + Sys + Wash + Dry + Dip + Spray + Rinse + Wash(Rinse) + Dip(Rinse) + Spray(Rinse) + Rinse(Wash + Dry) + Rinse(Wash + Dry + Dip) + Rinse(Wash + Mlk + Mdim + Mlk + Mdim + +8 Mlk + Mdim + Mlk + Mdim + Dry Lac Lac Lac Lac 4. + + + Dip(Btrt + Numtrt) Spray) + e Sys + Masmlk Sys + Lactrt + Trtad + Lactrt(Trtad) Sys + Numtrt + e Sys + Btrt + Numtrt + Dip + +8 Mlk + Mdim + Lac + Sys + Btrt + Numtrt + Bdip + Bdip(Btrt + Numtrt) + e Y8 All factors models. 63 Mlk + Mdim + Lac + Sys + Wash + Dry + Dip + Rinse + Numtrt + wash(Numtrt) + Dip(Numtrt) + Numtrt(Wash + Dry) + Numtrt(Wash + Dip) + Numtrt(Wash + Dry + Dip) + e Mlk + Mdim + Lac + House + Bed + Bedadd + Bed(Bedadd) + House(Bed + Bedadd) + e Mlk + Mdim + Lac + House + Bed + Bedrake + Bed(Bedrake) + House(Bed + Bedrake) + e Mlk + Mdim + Lac + House + Bed + Bedclean + Bed(Bedclean) + House(Bed + Bedclean) + e Mlk + Mdim + Lac + House + Bed + Bedadd + Bedrake + Bedclean + e Mlk + Mdim + Lac + House + Ovsc + e except error were fixed in each of these management practice O RESULTS AND DISCUSSION Effect of various cow factors on somatic cell count A model containing herd, cow within herd, daily milk production (lbs/day), parity, cow age, stage of lactation, percent milk fat, herd size, and month of year was constructed to examine how cow related factors affected somatic cell count (SCC). Actual SCC divided by 100,000 plus one was translated into natural log (1n SCC). Analysis was based on 1,073,587 cow records in 1,109 herds. The model accounted for 43.35% of variation in 1n SCC. The resulting analysis of correlation summary table is presented in Table 5. All variables in the model had a statistically significant impact. Cow within herd accounted for 64.01% of model variation, herd 21.06%, percent milk fat 6.36%, month of year 4.38%, parity 3.71%, daily milk production 0.23%, stage of lactation 0.23%, age of cow 0.11% and herd size .00074%. The large percentage for cow within herd variation in this analysis is primarily due to the large variation in farm conditions and managerial practices between herds. Further analysis of individual effect of above listed factors on somatic cell count was performed. Variance due to herd and cow within herd was removed. As daily milk production increased, ln SCC decreased, and as parity, age, stage of lactation, percent fat and herd size increased, 1n SCC increased. 64 65 —mp.moo.omw 5mm.mmo.p peach oouomgeou .em.mmm.pm¢ map.omm totem po. oeu.~ — oNFm use: .ooo. ewp.~¢ _ ma< zoo .coo. ~m~.¢m¢.m~ _ one x—wz & .ooo. om¢.wma.o— .p coco: .ooo. Nmm._mm _ .u<4 macaw —ooo. m¢~.muc.m~ Np suntaa —ooo. m.m.mom _ .cota x—_z _ooo. mum.omm.mm~ om_.o._ zoowetmz —ooo. ~__.moo.- mo_.P new: cmm.mee.mom om~.~_p _muoz Av av mm .m.u measom mueao_e_=m_m .sga553m :o_umpmttoo no mwmapmc< no epoch mtouomm zoouu.m mmm>02w ._<._.O._. 74 were obtained from five Michigan National Oceanic and Atmospheric Administration stations. In general, peaks in amount of precipitation corresponded to seasonal peaks in SCC. Simensen (1974) and Titterton and Oliver (1979) reported increased SCC with increased precipitation. The greater elevation of June SCC compared to October SCC may be an additive effect of higher environmental temperatures. Nelson et al. (1967), Wegner et al. (1976), Nelson et al. (1969), Whittlestone et a1. (1970), and Paape et a1. (1973) have shown elevated SCC during periods of warm environmental temperatures. Additionally, a change in cow's ration may also have contributed to SCC increase (Carroll, 1977; Ruffo and Sangiorgi, 1980). The secondary SCC peak in October may be the result of an interaction among atmospheric temperature, precipitation, and housing. Amount of precipitation declined from June to September, then increased (Figure 3). Increased precipitation could have increased SCC (Simensen, 1974; Titterton and Oliver, 1979). The combination of cooler temperatures and precipitation has been shown to have a chilling effect, different from cold temperatures alone, on the udder. The chilling effect has been associated with elevated SCC (Simensen, 1974). A housing and weather interaction has been shown to affect incidence of mastitis. Vasil (1980) reported increased clinical mastitis in cows housed in barns with poor ventilation and barn humidity of 80% or more during summer-fall weather transition. Frances et al. (1981) found that as environmental temperatures decreased cows spent 50% more time lying in free stalls which was associated with an increased probability of udder infection. In 75 addition, a shift in cows ration to more ensiled feeds may have affected mean SCC (Carroll, 1977; Ruffo and Sangiorgi, 1980). In addition to factors discussed here, other factors may affect somatic cell counts. These factors include, but are not exclusively limited to, date of freshening, heredity, wind chill, relative and absolute humidity and radiant heat (Lush, 1950; Afifi, 1967; Nelson et al., 1967; Afifi, l968a; Nelson et al., 1969; Wilton et al., 1972; Simensen, 1976). Effect of milking and housing systems on somatic cell count Milking System.--Improper function or use of the milking system has been suggested as preconditioning the cow to teat injury, to increased incidence and spread of mastitis, and to higher somatic cell counts. It was the goal of this portion of the study to deter- a mine how various milking system factors affected SCC under field conditions. Mean values used for analysis of effect of milking system on SCC are presented in Table 6. Compared to herds with milking systems containing more than one brand of equipment Conde®, Bodmin®, and Boumatic® milking systems were associated with significantly (p < .002, .04, and .005, respec- tively) higher mean somatic cell counts. Germania® and Zero® milking systems had a negative association with mean SCC, but it was not significant (p < .40). There was no significant (p < .20) differences in mean SCC among other brands of milking equipment. Since none of the systems surveyed were inspected to assure correct installation and operation, it is not possible to definitely state why only the 8.8 888 88 8..8 8.88. 8.. 8.8. 8..8 8.8 ...8 888. 888.8 888. 888. 888.. 8 . 88. 8.8 888 88 8.88 8.8.. 8.8 8.8. ..88 8.8 8..8 888. 888.8 888. 8.8. 888 . 8 . 88. 8.8 .88 88 8..8 8.88. 8.8 ..8. 8.88 8.8 8..8 888. 888.8 888. .88. 88... 8 . 88. ..8 888 88. 8.8. 8.88. 8.8 8.88 8.88 8.8 8..8 888. 888.. 888. 888. 888.. 8 . 88. ..8 8.8 88. 8.8. 8.88. 8.8 8..8 8..8 8.. 8.88 .88. 88... 8.8. .88. 888. 8 . 88. ..8 ..8 88 8.88 8.8.. ..8 8.88 8.8. 8.8 8.88 888. 888.. .88. 888. 8.8. . . 88. 88.. 88. .88—888 ..8. 888 8. 8.88 8.88. 8.8 8.8. 8.8 ..8. 8..8 8... 888.8 888. 88.. .88.. 8 . 88. 8.8 8.8 .. 8.8. 8.8.. 8.8 8.8. 8... ..8. ...8 888. 888.8 8.8 888. 888.. 8 . 88. 8.8 888 8. ..88 ..88. 8.. 8..8 ..8 8.8. 8.88 8.8. 888.8 8.8 8.8. 888.. . . 88. 8.8 .88 88 8.88 8.8.. 8.. 8..8 8.8. ..8. 8.88 888. 888.8 8.8. 888. 888.. 8 . 88. 8.8 ..8 8. 8.88 8.88. 8.. 8.8. 8..8 8... 8..8 888. 888.8 888. .88. 888.. 8 . 88. 8.8 888 88 8.88 8..8. ... ..8. 8.88 8.8 ...8 ..8. 8.8.8 888. 888. 888.. 8 . 88. 8.8 888 88 ..88 8..8. ... 8... 8..8 8.8 8..8 888. 888.. 888. 8.8. 88... 8 . 88. 8.8 888 8. 8.88 8..8. 8.. 8.88 8.88 8.8 8..8 ..8. 888.. 8.8. 8.8. 888. 8 . 88. ..8 888 .8 ...8 ..8.. 8.8 8.88 8.88 8.. 8.8. 888. 888.. ..8. 8.8. 8.8. . . 88. 88.. 88.8 .88._oa.8 8.88 -- 88 88 -- ..88 -- ..8. ..8. -- ..8. -- 888.. 888.. -- 888.. 8.. 88. -- 888 888 -- ...88 -- 8.88 8.88 -- 8.88 -- 8...8 8.. 8 -- 8.. 8 8 . 88. -- 8.. 8.. -- 8.8.. -- ..8. ..8. -- ..8. -- .8... .8... -- .8... 8 . 88. -- 88. 88. -- 8.88. -- ..88 ..88 -- ..88 -- 888.. 888.. -- 888.. 8 . 88. -- 8.8 8.8 -- 8.8.8 -- ..88 ...8 -- ..88 -- .88. .88. -- .88. 8 . 88. -- 888 888 -- 8..88 -- 8..8 8..8 -- 8..8 -- .8... ..... -- ..... 8 . 88. -- 888 888 -- 8.8.8 -- 8.88 8.88 -- 8.88 -- 888. 888. -- 888. . . 88. 88.. 38. ..8.—88.8 8.88 8.8. 8.8 88 8.88 8.88. 8.8 8... 8.8 . 8.8. 8.88 8... 888.8 888. 888. 888.. 8 A 88. 8.8. 888 88 8.8. 8.88. 8.8 8..8 8.8 8.8. ..88 8... 888.8 888. 888. 888.. 8 . 88. ..8. 888 .8 ..88 8.88. ..8 ..8. 8.8. 8.8. 8.88 888. 888.8 888. .88. 8.8.. . . 88. 8.. 888 8 8.88 8.88. 8.. 8.88 8.88 8.8 ..8. 888. .88.. .88. 888. 888.. 8 . 88. 8.8 888 88 ..88 8 88. ... ..8. ..88 8... 8.88 888. 888.8 8.8. 8.8. 8.8.. 8 . 88. 8.8 8.8 88 8.8. ..88. 8.. ..8. 8... 8... 8.88 888. 888.8 888. 888. 8.8.. 8 . 88. ..8 .88 8. 8.8. 8.88. 8.. 8.88 8.88 8.8 8.88 888. 888.8 .88. 8.8. 88... 8 . 88. ..8 888 .8. 8.88 8.88. 8.. ..8. ..88 8.8 8.88 .88. .88.8 888. .88. .88.. 8 . 88. 8.8 888 8.. 8.88 8.88. 8.. ..88 8.88 8.8 8.88 888. 888.. 888. 888. 888. . . 88. .88888 -h8pm.-x:umwzl .888-:;:.h8umxnllxmmmx “8.8 .888 .8.z.:::n8u81.ylmmmx .8-8 .8.: .8.8l .8.8 8.88 88.8.8 ...: ...: 8. 88.8 8.8: .88..:8888 ...: 8.8: 888 8. 8.8: ll"- 3 Ill-l nllaltl 14'1" I- 0.. I... Oli‘1313 {OI-V. ’]-|.11--.I|‘"l‘-‘ll‘1 I+O| I‘Ial‘l.‘ "lull' - I 1|! ‘ "’I.IO-‘.‘J‘\. O 1 1-11'.-.‘ 1' .99m :0 ......“me 6:.J—az so «0055u ha m—mh—CE has an: mos—Q) =8:!-.0 w3 on mean SCC. Significantly (p < .005) higher mean SCC associated with Surge VSO® take-offs are probably associated with one or more of the following: 1) claw weight, 2) lack of support for claw, and/or 3) a possible interaction of inflations being used and Surge VSO®. O'Shea et al. (1979) reported 50-60% of new infections were due to liner slip. However, more research is needed to determine the validity of each of these hypothese. Housing System.--Limited work has been done on how housing system affects milk somatic cell counts. The first part of this study examined the effect of milking herd housing system, bedding, and possible housing-bedding interactions on milk mean SCC. Mean values used for analysis of effect of housing system on SCC are presented in Table 7. Sand based free stall housing for the milking herd was associated with significantly (range p < .02 - .0008) lower mean SCC than any other housing system. There were no significant (p < .20) differences in mean SCC among the other housing systems. However, when ranking housing systems from smallest to largest 86 o... .vm mm a... o.vm. n.n ..we a... o.n. ..vc cm.. mo..~ m... ~c.. .9... m a o<. m.m. ~.v ow v.nm a.wm. ..o m.ae. ..~ o.a~ n.3m a... eme.~ o.c mac. n-.. a - o<. m.o. awn mo v..o v.~m. ..~ ...m m.a. o.m. a.cm «.3. mmq.~ .mc. n... vo~.. . u u<. m.w am. as o.o~ a.am. ... m.a. m..n o.o. n.cm moo. awn.~ .vm. o.m. a-.. a n u<. m.v can .a n.a~ m.mw. e.. ~.oo ..wm ~.a m.om .co. ..~.~ new. mmm. we... m u u<. ..v cam mo n.n~ ..co. o.. c..c a... n.m ..mc ans. aem.~ moo. men. co... c a q<. o.c .mw ... ...m n.n.. ~.. v.mm ..mm o.. c.me one. ecm.~ cam. own. emc.. n u u<. e.e mam a.. ¢.o~ n.ac. n.. ~.=o ..ew m.. m..v one. .em.~ c.m. can. .ma. ~ u u<. e.c com um. o.o~ ..m.. o.. ..na c..~ ~.c ..Nc mes. cea.. own. now. n... . u u<. m..uum out. ouueoemu >..o.=p ..n. eNN an. n.on ..mm. ..m v.vo o.om ¢.. “.mv new. :ca.~ ~.m.. ova. cam.. a . ac. o.m. .NN mm m.n¢ ..cq. n.m o.~o c.= a.n~ a..v mum. .n~.~ c.o. .m.. cmm.. a n u<. m..~ new no m.m. ..m:~ N.v ~.=a ..n. N... m.mv mam. o~=.~ .am. com. com.. . u u<. m.o. m.~ on. m..~ ~.nm. m.~ m.na c.am ..o m..« men. ~m..~ .No. new. Nem.. o a u<. o.~. .am e. m... m.mv. v.~ ~.no n.am ... a..m m7.. ace.~ o.o.. mvc. mam.. m - u<. m.~. mcm ... m.~c m.n.. e.~ m.oe n.~m m.. ..m. m... :m:.. mvu. cow. No... v a o<. ..m mm. as ..om m.nw. m.. ..m. ~.ov ..m m..c m=.. mca.. vwo. non. aw... n - u<. ... o.~ em. m.a~ ~.cc. o.. ~.Nm e.on n.m m.ee one. .cm.. ~¢o. ecu. mac.. u - u<. o.. n.~ .m. ~.o~ o.... m.. m..v ...m o.m o.cq mos. oo~.. mom. o..~ omm. . a u<. m..a.m out. .emeou ..uz .... mm. on m.oe o.oM. o.v o.m. =.n. m.m. a..m em.. .uw.~ mac. mac. m.o.. a - u<. c.e. .ou ac n.mm ...m. m.v v.v. ..c. o.a. ~.mq .... o.=.~ o.a m... can.. a n ua.u m.ov won a. ~.oc. m.... m.m c.:. ...m n.n. o..e mus. m:o.q .vm. emm.. .n... a s u<. m.a mm. m. n.n~ ..no. ~.n ..ac a..m c.. o.me .em. aeo.~ mmv. ems. ~mm.. m u u<. m... ~v~ mo. o.mv ..~.. a.~ q.on a..m 0.: o..« a... aam.~ .oc. amo. mmm.. . u u<. a... ..~ on n.ov o.mm. m.~ ~.=n n.m~ ..a v.mv .... ..o.~ ~.o.. .eu. amm.. o u u<. ~.o~ .nN ow o.¢o ~.cc. o.~ ~.no c.mn ..a ...v an.. ace.. 0.: ~3n. o«~.. m a u<. .... new as ..mn «.mm. ..N ~.=o c.en ¢.a m..v .n.. c.~.~ vac. vac. .c~.. v u u<. a.m mo~ co. m.a~ ¢.co. m.~ o..o ..Nn ~.m v.co ~;:. mmn.. .mc. mow. moo.. n n u<. m.o mm. o~. c.n~ o..m. o.~ a.~n ...n m.e ..mo .oc. cam.. now. .-. «so. ~ - u<. ~.¢ ~.~ oN. n..~ 0.... ... ..Nv ..aN o.m c..n 23:. m~o.. oov. ~o.. com. . a u<. new. esotu . .m .anI .e.x .c.m can: ...m .xa: .=.z Harm..--emux Mumm: I.uxa: .=.:a|::hcmmnz:szwmx iguana v i ”am ...... 5...: ...: e. «as: can: -ll‘l. u:ma...=1at. ...z.=euz Avvac.ucouv . u.m<. 88 effect on increasing mean SCC, totally cemented free stalls had the smallest effect, followed by stanchions, group pens, year round pasture, comfort stalls, clay based free stalls, and half cemented free stalls. Bedding type by itself did not have a significant (p < .20) on mean SCC. However there were several significant housing systemebedding type interactions. In comfort stall housing systems, use of both long straw and sawdust as bedding was associated with significantly (range p < .06 - .0001) lower mean SCC than any other bedding materials. Though not statistically different, the ranking of other comfort stall beddings from smallest to largest numerical effect on mean SCC was sand, chopped straw, long straw and sawdust. In half cement free stalls (front half cemented, rear half dirt base), bedding with a combination of long straw and sand was associated with significantly (p < .05) lower mean SCC than bedding with sawdust alone. This finding is not surprising for two reasons. First, sawdust bedding has been directly linked to an increased coliform infection rate and clinical mastitis (Newman, 1973; Bramley, 1974; Newman, 1975). Second, due to their design, half cement free stalls would tend to collect moisture and feces in the back half of the stall which is conducive to the growth of microorganisms. Use of both long straw and sand in half cement free stalls probably leads to formation of a relatively dry manure pack and thus lower mean SCC. In totally cemented free stall housing systems, lower mean SCC was associated with use of corn cobs/stalks, sand and sawdust, hay and sand, or chopped straw and sawdust. There were no significant (p < .20) differences in mean SCC among these beddings. 89 However, mean SCC in housing systems with cemented free stalls using sand bedding alone or long straw bedding alone was significantly (p < .05 and .002 respectively) higher. These results may be related more to management factors; such as depth of bedding, frequency of bedding addition, raking, or stall cleaning and weather conditions rather than actual bedding material (Ekesbo, 1966; Francis et al., 1981). In group pen housing systems, mean SCC was significantly (p < .002) higher when long straw bedding was used than when the combination of long straw and corn cobs/corn stalks was used as bedding. There was no significant (p < .20) difference in mean SCC between use of chopped straw and either long straw or corn cobs/corn stalks plus long straw in group pens. Again these differences in mean SCC may be the result of management practices or weather conditions and not the bedding material per se. The second part of this study examined maternity facilities; type, number of stalls if applicable, type of bedding, type of facility-bedding type interaction, and seasonal interactions. A significant decrease in mean somatic cell count was associated with the presence of maternity (box) stalls. Herds with four to seven maternity stalls had significantly (p < .10) lower mean SCC than herds with one or two maternity stalls. Herds with three maternity stalls were not significantly (p < .20) different from herds with either one or two, or four or more maternity stalls. Lowest numeric mean SCC was associated with five maternity stalls. When type of bedding in maternity stall(s) was considered, a significant (p < .01) reduction in mean SCC was associated from 90 greatest to least, with two maternity stalls bedded with chopped straw, two maternity stalls bedded with long straw, three maternity stalls bedded with long straw, and one maternity stall bedded with long straw. There were no significant (p < .20) differences among these four types of calving facilities. Effect of season at calving and location of majority of calvings on mean SCC was examined. Significantly lower mean SCC were found when, during fall to spring calving period, a majority of calvings occurred either in out-of-door pens with no shelter or in pasture (p < .0002 and .03). Out-of-door pens with no shelter were associated with significantly (p < .05) lower mean SCC than pasture. Significantly (p < .01) higher mean SCC were associated with herds where the majority of fall to spring calvings occurred in an out-of-door pen with shelter. During spring to fall calving period, numerically lower mean SCC were associated with herds where majority of calvings occurred in either box stalls or group pens. Mean SCC was significantly (p < .05) higher in herds where the majority of calvings occurred either in the cow's normal stall or where ever the cow was at time of parturition. Further analysis was performed to determine the interaction effect, if any, of season of calving, location of calving, number of maternity stalls, if applicable, and maternity stall bedding on mean somatic cell count. Those conditions associated with signif- icantly lower mean SCC during fall to spring calving period are listed in Table 8 (p < .05), while those associated with the spring to fall calving period are listed in Table 9 (p < .10). Only significantly 91 omm.Nn zegom condone . can macaw mNm.N- sebum ago. - two.m:m no.3 :ma-.oou-.o-u=o eeo.m- zetum mco. . ..mum xon emo.mu m..mum\mnoo ctoo . ton.m;m se.; cma Loco-.onu:o mN..m- sebum omaaozo N ..oum xoa cc..m- xenon use. N ..mum xoa mo..m- sebum mac. . cog esotm NON.N- >8; - can esotm eON.m- zmtom ago. . ..mum xon 5.? :2: 95. N :3 95.5 ¢.N.m- sebum umaaogu - :ma toov-.o-u=o ..8...- 32.... 93. m 23.. xon eNm.m- omsu 38m . ..mum xon mme.m- sebum meo. . can macaw .uum came co .um..m .omm. .a.tmamz m..oom >u.=goumz >u...ucm 35.3.8 8°60. :5 m 9:25.. $4.52 -|~ Ill .uum cam: go mco.s.s=ou m=_>.mu m:..am o. .... .o somu.m--.m msm<. 92 .om..- seenm ago. m ..msm xos ¢.o.N- zatum mco. . can esotm mm..N- sebum ago. N ..mom xon mNm.N- zetom mac. . ..eum xon m...N- wee: . can anotm mNm.N- umaczam N tma.m;m se.: con toouumouozo .em.m- wee: - twu.m;m no.3 :wa tooca.ouu:o mam.m- .msuzmm . ..aum xon .uum game go sum..m .smo. .a.tmsoz m..esm .u.=tmu~= as...omi ..e\m..mu ooo.oo. c.. m m:.uumm amassz “H Ill-9|- I'll .oom :mwz :o mcomawocou m=.>.co ..8. o» mcwgam mo uuowmmnn.m m4mm specs tweuo som>o 3mm; a maze gazes a muco gem; tmzuo .um>o emu: a mono some tango >gm>m gatom tonnage amatztm serum mco. umsuxnm “museum teem sebum umaaoeu tmacmsmu ..m: toucmsmo x..muoe tenemEmu x..muoe nonemEmu a..euo. umueoeoo A..mpo. umucmEmu A..muo. amazoEmo a..muo. .uum came :6 uuo..m .umm. ..exm..mu ooo.oo. :.v m a=_uomm to =o_s_ue< ...tmssz s=_ssmm m:.m:o: ..mum amt. '1!" .uum com: co a=.eumm to =o_s.ts< use ....tmsmz m=_esmm .smsm.w m=_m=o: ...sm mat. to som..u--.o. msme. 104 somatic cell count. In numerical order, least positive effect on mean SCC to greatest, these were totally cemented free stalls bedded with chopped straw and raked every day, totally cemented free stalls bedded with sawdust and raked every day, and half cemented free stalls bedded with sawdust and raked every other week. It is not possible to satisfactorily explain these results at present. Perhaps raking is influenced by other factor(s) not included in the current model, such as addition of bedding, thoroughness of raking, etc. There were two significant (p < .05) free stall, bedding material, and cleaning free stall down to base and leveling inter- actions on mean somatic cell count. Totally cemented free stalls bedded with sawdust and cleaned and leveled once a month were associated with significantly (p < .05) lower mean SCC. Signifi- cantly (p < .05) higher mean SCC was associated with totally cemented free stalls bedded with sawdust and cleaned and leveled once a year. Francis et al. (1981) reported increased levels of §h_g21i_in sawdust bedded free stalls as environmental temperatures decreased and cows spent more time lying in free stalls. Additionally, §h_ggli levels were higher in free stalls of high producing cows as compared to low producing cows. Thus, more frequent cleaning probably reduces the number of pathogens below critical levels. Due to the size of the data bank, it was not possible to analyze for interaction effects of addition of bedding, raking, and/or cleaning-leveling of free stalls on mean somatic cell count. 105 Purchase of Replacement Animals.--Herds that had purchased any lactating cows in the past two years had significantly (p < .0009) higher mean somatic cell counts than herds that had not made any purchases of lactating cows. Considering the high probability of purchasing an infected lactating cow; eSpecially, if the purchased cow does not have a complete health record or is not tested for mastitis at purchase, the above results are as expected. GENERAL DISCUSSION Somatic cell count has been determined to be a reliable indicator of mastitis (Waite and Blackburn, 1957; Schalm and Lasmanis, 1968; Schalm et al., 1971; Reichmuth et al., 1974; Reichmuth, 1975; Westgarth, 1975; Bodoh et al., 1976; Schultz, 1977; NMC, 1978; Eberhart, 1979; McDermott et al., 1981; Jones et al., 1982). However, SCC has been shown to be influenced by factors other than infection. This study was designed to examine the effect of cow related, farm, and managerial factors on SCC. Cow factors affecting somatic cell count can be divided into three categories: 1) physiological factors, 2) lactation product factors, and 3) environmental factors. SCC increased as physiological factors of age, parity, and stage of lactation increased. This increase is probably due to increased exposure to infection and/or increased incidence of subclinical mastitis as each of the phys- iological factors increase. Examination of lactation product factors found as SCC increased, daily milk production (yield) decreased and percent fat increased. The decrease in daily milk production is due to damage of mammary cells, the stimuli which is causing increased SCC, while increase in percent fat is a concentration effect. While SCC increased as environmental factor of herd size increased, this increase was not biologically significant. Analysis of seasonal trends in SCC found a yearly peak in SCC occurred in June and a 106 107 secondary peak occurred in October. While these peaks are probably the result of interaction of several factors, environmental temperature and precipitation most likely account for majority of the observed phenomena. However more detailed and controlled research is needed. Farm factors possibly affecting somatic cell count were divided into two categories: 1) milking system and 2) housing system. Only high milk pipeline milking parlor milking systems were associated with significantly lower mean SCC. This result is different from previous reports of low milk pipeline parlors having lowest SCC (Downey et al., 1977; Schultz, 1977). However, LeDu (1980) reported that with a low milk pipeline milking system, more vacuum fluctuations occurred during milking, especially, at end of milking, and transmis- sion of vacuum fluctuations from a cluster to the nearest cluster occurred more easily. Lower mean SCC were also associated with line vacuum levels of 11.5 inches Hg and 13.5 inches Hg for low and high milk pipeline systems, respectively, complete loop vacuum line design, and single pulsation. Pulsation ratio of 50:50 was associated with lowest mean SCC over widest vacuum level range. This is probably because 50:50 ratio provides adequate massage and thus, there is minimal teat injury and potentially less spread of infection. Pulsation ratios of 55:45 and 60:40 were also associated with low cell counts over a wide range of vacuum levels, but not as wide a range as 50:50 pulsation ratio. Investigation of interaction of milking system and vacuum controller on mean SCC found significant interactions of brand of controller and barn pipeline and parlor milking systems. Why only certain brands of vacuum controller had 108 an effect is probably due to load change in weighted sleeve value and weighted level controllers which are less sensitive to vacuum fluctuations, while installation, lactation, and maintenance practices have a greater effect on the more sensitive spring loaded diaphragm controllers than load charge. Lowest SCC were associated with five milking units on 1%, 2, or 2% inch milk pipelines and eight milking units on 3 inch milk pipelines. Brand of inflation being used probably has an effect on SCC, but controlled studies are needed to determine the effect of each brand of inflation on mean SCC. Mean SCC was not affected by the presence or absence of automation, unless Surge VSO® take-offs were in use. Higher mean SCC associated with Surge VSO® may be because of one or more of the following: 1) claw weight, 2) lack of support for claw, and/or 3) possible interaction of inflations being used and Surge VSO®. Material of vacuum line construction, vacuum line size, vacuum pump capacity, and interactions among various pulsation ratios and single or alternate pulsation did not have a significant effect on mean SCC. How milking cow housing system and maternity facilities affected somatic cell count was examined. Lowest mean SCC were associated with sand based free stalls. There were significant interactions between certain housing systems and bedding materials. This indicates some housing system-bedding type interactions may keep bacterial numbers below critical infection levels, while others either enhance or promote bacteria growth. However more detailed controlled research is needed to support or refute this possibility. A general examination of maternity facilities found herds using 109 maternity or box stalls bedded with either chopped or long straw had lower mean SCC. When season of calving was also considered, different facilities were better during different seasons, but multiple calving facilities, one for each season, are not practical. There- fore, best year round calving facility appeared to be two maternity or box stalls bedded with long straw. These results confirm the importance of proper calving facilities as a mastitis control measure. Five areas directly controlled by the herd manager are milking hygiene practices, treatment of mastitic cows, dry cow treatment policy, free stall maintenance, and purchase of replacement animals. Lowest mean SCC were associated with the following milking hygiene practices: washing teats with individual paper towel using water containing a sanitizer, drying with a separate paper towel, teat dipping with an effective teat dip immediately after removal of milking machine, and rinsing or backflushing of teat cups between cows. Prepping should last 20-30 seconds and milking machine attached 30-60 seconds after completion of prepping. While we found herds milking mastitic cows first were associated with lowest mean SCC compared to other milking practices, these herds had either bucket milking systems or milking systems with jars, making it possible to totally clean the portion of the system that was in contact with infected milk, and only accounted for 0.72% of total population studied. In general, herds where mastitis was primarily treated by intravenous or intramuscular injection of antibiotics had higher mean SCC. These results are not surprising considering 110 successful treatment depends on effective passage of drug from site of injection to blood to foci of infection to milk (MacDiarmid, 1978; Ziv, 1980a; Ziv, 1980b). Dry treatment policy is important since the bovine udder is most susceptible to infection at the beginning of the dry period because bacteria are able to penetrate the streak canal more easily (Cousins et al., 1980) and administra- tion of dry treatment has been shown to reduce new infection rate and cure existing infections (Oliver et al., 1962; Natzke, 1971; Meaney and Nash, 1977; Rindsig et al., 1978). Lower mean SCC occurred in herds that dry treated all cows compared to any other dry treatment program. However, effectiveness of dry cow treatment policy was influenced not only by number of cows in herd treated, but by dry cow treatment product used, whether or not teat dipping was done during lactation, and brand of teat dipping product used, if any. The prime managerial concern with free stall maintenance is keeping bacterial numbers in bedding below critical infection level. Lowest mean SCC was associated with totally cemented free stalls, bedded with a fine textured bedding; i.e. sand, sawdust, chopped straw, etc., raked every day and cleaned down to the base once a month. Totally cemented free stalls facilitate removal of contam- inated bedding, especially fine textured bedding, because of its slope. Thus, reducing bacterial numbers. However Francis et al. (1981) has indicated weather may influence bacterial numbers in free stall bedding. Therefore more research should be done in the 111 area of free stall maintenance and factors affecting bacterial numbers in bedding. Since many of the lactating cows sold for dairy purposes, i.e. non-slaughter, are sold because of low production and/or mastitis, it is not surprising herds purchasing lactating cows have higher mean SCC than herds not making such purchases. Purchase of lactating cows probably involves unknowingly purchasing a new pathogen in a majority of cases. The new pathogen is then spread to other cows in the herd, elevating their cell counts. In conclusion, this study has shown somatic cell counts to be influenced by many other factors besides infection and/or injury. Fortunately, many of these factors can and should be controlled by the dairy farmer. These factors include proper operation and main- tenance of all components of the milking system, following proper milking hygiene practices, proper treatment of lactating cows with mastitis, dry treating all cows with an effective product at end of lactation, providing and maintaining adequate calving facilities and housing for lactating cows, and limited, if any, purchase of lactating cows as herd replacements. Additional practically oriented research is needed in many of these areas, especially in area of brand name comparisons. Further research is also needed in how environmental parameters; i.e. air temperature, precipitation, radiant heat, wind chill, etc., affect not only SCC, but other farm conditions, i.e. housing, bedding, etc. APPENDICES APPENDIX A SURVEY QUESTIONNAIRES .‘leLKING SYSTEM What type of milldn boxes). mu bucket barn pipeline parlor herringbone side opening rotary trigon polygon What make of equipment do you have? Surge Delaval Bou Ma tic How many milking units does this system have? Do you have weigh jars? l llll V35 no Delaval "300" Delaval "200" Delaval "Trade-offs“ Surge QTO Surge VSO Herd No. 3 system are you presently using? (please check the appropriate No. stalls No. stalls No. stalls No. stalls No. stalls ChoreBoy __ Other Universial-StaRite Zero Ccnde Bodrnin Do you have any of the following automation? Bounlatic "Take-offs" Ger-mania "Take-offs" Cniversal-StaHite "Take-offs" Other, Specify None :5 YOU HAVE ANY AUTOMATION IN YOUR SYSTEM, is it powered by HI the milking system's vacuum pump. a separate vacuum pump. air compressor 0 ther, please specify 9 How many inlets are there to the receiver jar? HI one two three or more ’4 [U 10. u. 16. 113 In order to move milk to the bulk tank, does your system have a milk pump on the receiver? releaser between the receiver and bulk tank? vacuumized bulk tank? milk transfer station (veyor)? other, please specify - What is the horsepower of the vacuum pump(s)?. What is the vacuum level of your system on the vacuum line or by the trap or receiver? inches What make of vacuum regulatods) do you have? Surge Oll Weight Westphalia (Vac-U-Rex) Surge (Equalizer) Sentinel Delaval (Senior) Bouhlatic, Universal, StaHite, Alta Laval (Servo) or Zero Balance Arm Germania ChoreBoy ‘ Other, please specify How many vacuum regulators does your system have? Where are the vacuum regulators located? (check more than one if more than one location) on the balance tank on the line between the balance tank and the receiver on the pulsation line by the moisture trapon the receiver , by the wash trap other, please specify What is the size (diameter) of the pulsation or vacuum line? 3/4 inch 1 1/4 inches 1 1/2 inches 2 inches 3 inches The pulsation or vacuum line is galvanized pipe. plastic (PVC). The pulsation or vacuum line is a dead end. looped 1'. complete loop back to the balance tank. 13. 2°. «.60 ..e f‘ .5. What is the pulsation ratio of your equipment? 50:50 55:45 60:40 65:35 70:30 50:50 front, 80:40 rear 5‘ the pulsation alternating (two teats at a time) or single (all four teats at the same time)? ' alternating single How often is the pulsation or vacuum line cleaned? after each milking once every six months once a month once a year once every three months has never been cleaned other, please specify What is the height of the milk pipeline? below the cow's udder 3 to 5 feet above the cow's udder 5 to 8 feet above the cow's udder more than 8 feet above the cow's udder What is the size (diameter) of the milk line? 1 1/2 inches 2 inches 2 1/2 inches 3 inches What make of claw do you have? DeLaval Bodmin BcuMatic Surge Universal Germania Zero ChoreBoy Sta-Rite Other, please specify What is t he claw made of? plastic stainless steel What style of inflation do you have? (for example; Delaval Oi, Bouhlatic 3.1) 28. A0 V e Who is demanufacturer or source of your inflations? (for example, MMPA, Macs, Delaval) How many days are inflations used before they are discarded? days. Are sets of inflations washed, rested, and reused? used continuously before discarding? How often is the system serviced by a qualified equipment specialist or representative? more than twice a year twice a year ‘ once a year never other, please specify Are these calls mainly service contract? emergency service? Who do you consult when you want information on milking systems? if more than one, piease rank in order of importance with l 8 first. county agent equipment company advertising dairy extension specialist friends and neighbors dairy flieldman regulatory officials equipment company representative veterinarian other f-ias the milking system been changed in the last year? yes no if" YES, what was the nature of this renovation? Herd No. MILKING PRACTICES l. 3. 4. How many persons are involved in milking the herd? What is the position of the milker? (if more than one milker, please indicate how many persons are at each position.) owner-opera tor herdsman full time hired employee part time hired employee family member of owner-operator other, please specify If the milker(s) are employees, does that person have any additional non-milking related responsibilities? yes no Do you (please check) use a prep stall prep at cowside no prepping Which of the following items are part of your normal milking procedure? (please check the items which apply) NORMAL PREPPING PROCEDURE wash udder with common sponge or rag before milking wash udder with paper towel(s) before milking spray udder with running water before milking include a sanitizer in udder wash or spray NAME SANl'l'lZi-ZR dry udder after washing NORMAL PRE-MILKING PROCEDURE use a strip cup before machine attachment do a California Mastitis Test (GMT) on each cow before milking do a GMT on cows with abnormal milk before milking NORMAL POST-MILKING PROCEDURE machine strip less than 1 minute before removing milkers machine strip more than 1 minute before removing milkers dip teats after milking NAME OF TEAT DIP spray teats after milking NAME OF THAT SPRAY rinse teat cups in sanitizer between cows 6. 9. 117 Total time spent in prepping each cow is: less than it! sec. 20 to 30 sec. 10 to 20 sec. , more than 30 sec. Amount of time between prepping and attachment of milking unit is: less than 30 sec. 1 to 2 minutes 30 to 60 sec. more than 2 minutes Total milking time for herd: hours. in your herd, cows with mastitis and treated cows are milked (please check one). before the rest of the herd with the rest of the herd. . after the rest of the herd on special equipment not part of the regular milking system by hand not at all How would you rate your herd's teat end condition? acceptable fair eroded Who do you consult when you want information on milking practices? if more than one, please rank in order of consultation, = first. county agent commercial companies dairy extension specialist dairy field man friends, neighbors state sanitarian veterinarian other ..- .4 0) Herd No. intxmc HERD name amiss? rouc‘: 1. What is the dominate breed of dairy cattle on your farm? Holstein Jersey Brown Swiss Ayshire Other, please specify 2. How many total cows (milking and dry) do you have? cows ... How many cows are you currently milking? cows 4. As of your last DHIA report, what was your rolling herd average? lbs. 3. is your herd split according to production levels? Yes No 1? YES, HOW 15 THE HERD SPLIT? 6. Milking cows are housed in stanchions with dimensions of . comfort stalls with dimensions of half cement floor free stalls with difiensxons oT totally cemented floor free stalls with dimensions of group penis), manure pack, with square iEet per cow. pasture in summer. [F YOU HAVE FREE STALLS, PLEASE ANSWER THE FOLLOWING QUESTIONS. [F NOT, PLEASE GO TO QUESTXON 10. 7. Plow often is bedding added to the free stalls? once a week once every other week once a month once every other month once every three to four months other, please specify 8. How often are the free stalls cleaned down to the base of the stall and the base then leveled off? once a month once every other month once every three months once every. four months once every six months once a year other, please specify 10. £19 How often are the free stalls raked? every day every other day every three days once a week once every other week other, please specify What type of bedding are the milking cows on? llll sand sawdust chopped straw long straw corn cobs or corn stalks My other, please specify Have you purchased any lactating cows in the past two years? Yes We [F YES, were they checked for mastitis? Yes No IF YOU CHECKED A NEWLY PURCHASED COW FOR MASTITIS, HOW DID YOU DO ”i! ll? l3. 14. When a cow has clinical mastitis (abnormal udder and/or milk), when is she treated? immediately after milking after consultation with a veterinarian after the results of a milk culture sample are received other time, please specify Who treats a cow if she has clinical mastitis? owner-operator family member herdsman employee (milker) veterinarian other, please specify 15. 16. 17. 13. 20. (5 do [‘3 l9 L20 When a cow has clinical mastitis, what is used to treat the cow? individually packaged commercial syringes or canulas. individually packaged syringes or canulas prepared by a veterinarian. individually packaged syringes or canulas containing an autogenous vaccine developed especially for your herd by a veterinarian. common syringe and multiple dose commercial drug. common syringe and multiple dose drug mixed by a veterinarian. common syringe and multiple dose autogenous vaccine developed especially for your herd by a veterinarian. other, please specify . Where is the treatment for a clinically infected cow administered? intramammary intramuscularly sub-cutaneously intravenously in traperitoneal dry treat PO no cows. all cows. less than half the herd. more than half the herd. only cows that had mastitis that lactation. lF YOU ARE DOING ANY DRY TREATING, what product are you using? Do you keep a record of mastitis cases for all cows? some cows? only problem cows? no cows? Do you keep a record of the types of mastitis treatment used? Yes No Sometimes How many cows were treated for mastitis during the last twelve months? cows. How many cows were culled primarily because of mastitis during the last twelve months? cows. 121 23. Have you made any changes in your nerd management policy, herd housing facilities or herd size in the last two years? Yes No ii" THERE HAVE BEEN CHANGES, PLEASE BRIEFLY OUTLL‘iE THOSE CHANGES? l. 122 Herd No. DRY COW AND CALVING PRACTICES Are the dry cows separated from the milking herd? Yes No Dry cows are housed in Hlll stanchions with dimensions of . comfort stalls with dimension of . half cement floor free stalls with dimensTons of . totally cemented floor free stalls with dimensions of . group pen, manure pack, with square feet per cow. pasture in summer. What type of bedding are dry cows on? sand sawdust long straw chopped straw corn cob: or corn stalks hay other, please specify Do 2 year old cows calve in same maternity stalls or area as mature cows? Yes No During the fall'to sor'n; period of the year, most calvings occur (please check one) lHll in comfort or box stalls in a barn. in the cow's normal stall. in the cow's group pen. in an out-of-door pen with shelter. in an outwf-door pen with no shelter. in a pasture or field. where ever the cow is a parturition. other, please specify During the Spring to fall period of the year, most calvings occur (please check one) lllllll in comfort or box stalls in a barn. in the cow's normal stall. in the cow's group pen. in an out-of-door pen with shelter. in an out-of-door pen with no shelter. in a pasture or field. where ever the cow is at parturition. other, please specify 9. if you use box stalls for calving purposes, how many box stalls do you have? The bedding in the confined maternity area is corn cobs or corn stalks. hay. chopped straw. long straw. sand- sawdust. wood chips. none. other, please specify When are calves removed from the cow? Hill immediately after birth within 2 hours after birth 2 to 6 hours after birth 6 to l2 hours after birth 12 to 24 hours after birth 24 to 36 hours after birth other, please specify Are the calves fed fresh milk from mastitic cows? fermented milk from mastitic cows? no .7.lk from mastitic cows? Calves (birth to 8 weeks) are housed in Ill Who do you consult when you want information on dry cow and calving practices? individual calf hutches. individual pens or stalls in a barn. group per. in a cold barn. group pens in warm barn. group pens out--sf~icor3 with shelter if more than one, please rank in order of consultation, 1 = first. calvin L county agent commercial companies dairy estension specialist dairy fieldman friends/neighbors llll Hlllli veterinarian other Have there been any changes in either dry cow or calving facilities or practices in the last year? No Yes, please list the changes drv cow APPENDIX B PACKAGE GIVEN TO DHI TECHNICIANS CONTENT: Cover Letter Copy of Letter Being Mailed to Dairy Farmers Survey Questionnaires (See Appendix A) COOPl-ZRATEYE EXTENSION SER\'!C2£ \m mi.“ \l.-\l‘l-‘ l.'\l\‘l-RSIIY mi l' S. DH’ \R I \ll-V r (W \(‘nRICL‘ I. I L RI: COUPPIA “KG ‘ lMll'fl-Nf (I ”All? SCII'fiCI‘. . FAST LAW“ ‘ “KN‘AN ' "20 Dear DHIA Tester: .15 "E50 YOUR HELP.... We are trying to put together a comprehensive picture of the mastitis problem in Michigan. A part of this picture involves a survey of dairymen' 5 management practices and overall farm conditions. Using this information and DHIA production records, we will be able to determine what practices and on-farm conditions exist in Michigan and how they are related to somatic cell count, a reliable indicator of mast1.is. We will then be better able to lower the level of mast itis in Michigan dairy herds by identifbing those practices and conditions that lower the somatic cell count. In order to obtain as complete a picture as possible of Michigan dairy farms. we have developed four questionnaires covering different aspects of rarn operation. However to obtain any meaningful results. we need a response rate of at least 703 of the questionnaires. This is where your help is so critically needed. In the next three weeks, you will be receiving the first set of two questionnairs 1nd 1 cover letter explaining them. We ask that you take five to ten minutes .1th each lai1rxman you visit who is on the somatic cell count program, and ncl p hin oompl etc the questionnaires as accurately as poss ‘1ble. Afte the suest1‘onnaires are completed, just mail them back to MSU in the envelope that will be pr v1ded. The second set of questionnaires will arrive approxim tely three weeks after the first set, and we ask that you ahain help the dairyman sample: the questionnaires as accurately as possible. The :uestisr naires will not be a "pop quit" in the dair"man' 3 eyes. In 'pproVitttelr two weeks. ‘lichigan dairymen .ill be receiving a letter e: