Advertisement

The economic cost of metritis in dairy herds

  • Author Footnotes
    * Current address: Escuela Medicina Veterinaria, Universidad Autónoma de Santo Domingo, Santo Domingo, República Dominicana 10105.
    J. Pérez-Báez
    Footnotes
    * Current address: Escuela Medicina Veterinaria, Universidad Autónoma de Santo Domingo, Santo Domingo, República Dominicana 10105.
    Affiliations
    Department of Large Animal Clinical Sciences and D. H. Barron Reproductive and Perinatal Biology Research Program, University of Florida, Gainesville 32610
    Search for articles by this author
  • Author Footnotes
    † Current address: Department of Population, Health and Pathobiology, North Carolina State University, Raleigh, NC 27695.
    T.V. Silva
    Footnotes
    † Current address: Department of Population, Health and Pathobiology, North Carolina State University, Raleigh, NC 27695.
    Affiliations
    Department of Large Animal Clinical Sciences and D. H. Barron Reproductive and Perinatal Biology Research Program, University of Florida, Gainesville 32610
    Search for articles by this author
  • Author Footnotes
    ‡ Current address: College of Veterinary Medicine, Oklahoma State University, Stillwater, OK 74078.
    C.A. Risco
    Footnotes
    ‡ Current address: College of Veterinary Medicine, Oklahoma State University, Stillwater, OK 74078.
    Affiliations
    Department of Large Animal Clinical Sciences and D. H. Barron Reproductive and Perinatal Biology Research Program, University of Florida, Gainesville 32610
    Search for articles by this author
  • R.C. Chebel
    Affiliations
    Department of Large Animal Clinical Sciences and D. H. Barron Reproductive and Perinatal Biology Research Program, University of Florida, Gainesville 32610
    Search for articles by this author
  • F. Cunha
    Affiliations
    Department of Large Animal Clinical Sciences and D. H. Barron Reproductive and Perinatal Biology Research Program, University of Florida, Gainesville 32610
    Search for articles by this author
  • A. De Vries
    Affiliations
    Department of Animal Sciences, University of Florida, Gainesville 32610
    Search for articles by this author
  • J.E.P. Santos
    Affiliations
    Department of Animal Sciences, University of Florida, Gainesville 32610

    D. H. Barron Reproductive and Perinatal Biology Research Program, University of Florida, Gainesville 32610
    Search for articles by this author
  • Author Footnotes
    § Current address: Department of Population Health and Reproduction, University of California-Davis, Davis, CA 95616.
    F.S. Lima
    Footnotes
    § Current address: Department of Population Health and Reproduction, University of California-Davis, Davis, CA 95616.
    Affiliations
    Department of Veterinary Clinical Medicine, University of Illinois, Urbana 61802
    Search for articles by this author
  • P. Pinedo
    Affiliations
    Department of Animal Sciences, Colorado State University, Fort Collins 80521
    Search for articles by this author
  • G.M. Schuenemann
    Affiliations
    Department of Veterinary Preventive Medicine, The Ohio State University, Columbus 43210
    Search for articles by this author
  • R.C. Bicalho
    Affiliations
    College of Veterinary Medicine, Cornell University, Ithaca, NY 14850
    Search for articles by this author
  • R.O. Gilbert
    Affiliations
    School of Veterinary Medicine, Ross University, PO Box 334, Basseterre, St. Kitts, West Indies
    Search for articles by this author
  • S. Rodrigez-Zas
    Affiliations
    Department of Animal Sciences, University of Illinois, Urbana-Champaign 61802
    Search for articles by this author
  • C.M. Seabury
    Affiliations
    College of Veterinary Medicine, Texas A&M University, College Station 77843
    Search for articles by this author
  • G. Rosa
    Affiliations
    Department of Animal Sciences, University of Wisconsin, Madison 53706
    Search for articles by this author
  • W.W. Thatcher
    Affiliations
    Department of Animal Sciences, University of Florida, Gainesville 32610

    D. H. Barron Reproductive and Perinatal Biology Research Program, University of Florida, Gainesville 32610
    Search for articles by this author
  • K.N. Galvão
    Correspondence
    Corresponding author
    Affiliations
    Department of Large Animal Clinical Sciences and D. H. Barron Reproductive and Perinatal Biology Research Program, University of Florida, Gainesville 32610

    D. H. Barron Reproductive and Perinatal Biology Research Program, University of Florida, Gainesville 32610
    Search for articles by this author
  • Author Footnotes
    * Current address: Escuela Medicina Veterinaria, Universidad Autónoma de Santo Domingo, Santo Domingo, República Dominicana 10105.
    † Current address: Department of Population, Health and Pathobiology, North Carolina State University, Raleigh, NC 27695.
    ‡ Current address: College of Veterinary Medicine, Oklahoma State University, Stillwater, OK 74078.
    § Current address: Department of Population Health and Reproduction, University of California-Davis, Davis, CA 95616.
Open AccessPublished:January 14, 2021DOI:https://doi.org/10.3168/jds.2020-19125

      ABSTRACT

      The objective of this study was to estimate the cost of metritis in dairy herds. Data from 11,733 dairy cows from 16 different farms located in 4 different regions of the United States were compiled for up to 305 d in milk, and 11,581 cows (2,907 with and 8,674 without metritis) were used for this study. Metritis was defined as fetid, watery, red-brownish vaginal discharge that occurs ≤21 d in milk. Continuous outcomes such as 305-d milk production, milk sales ($/cow), cow sales ($/cow), metritis treatment costs ($/cow), replacement costs ($/cow), reproduction costs ($/cow), feeding costs ($/cow), and gross profit per cow ($/cow) were analyzed using mixed effect models using the MIXED procedure of SAS (SAS Institute Inc., Cary, NC). Gross profit was also compared using the Kruskal–Wallis test. Dichotomous outcomes such as pregnant and culling by 305 d in milk were analyzed using the GLIMMIX procedure of SAS. Time to pregnancy and culling were analyzed using the PHREG procedure of SAS. Models included the fixed effects of metritis, parity, and the interaction between metritis and parity, and farm as the random effect. Variables were considered significant when P ≤ 0.05. Metritis cost was calculated by subtracting the gross profit of cows with metritis from the gross profit of cows without metritis. A stochastic analysis was performed with 10,000 iterations using the observed results from each group. Milk yield and proportion of cows pregnant were lesser for cows with metritis than for cows without metritis, whereas the proportion of cows leaving the herd was greater for cows with metritis than for cows without metritis. Milk sales, feeding costs, residual cow value, and gross profit were lesser for cows with metritis than for cows without metritis. Cow sales and replacement costs were greater for cows with metritis than for cows without metritis. The mean cost of metritis from the study herds was $511 and the median was $398. The stochastic analysis showed that the mean cost of a case of metritis was $513, with 95% of the scenarios ranging from $240 to $884, and that milk price, treatment cost, replacement cost, and feed cost explained 59%, 19%, 12%, and 7%, respectively, of the total variation in cash flow differences. In conclusion, metritis caused large economic losses to dairy herds by decreasing milk production, reproduction, and survival in the herd.

      Key words

      INTRODUCTION

      Metritis is one of the disorders with greatest incidence during the early postpartum period in dairy cows. The incidence of metritis in dairy farms has been reported to be approximately 20%, but with a range from 8% to more than 50% in some herds (
      • LeBlanc S.J.
      Postpartum uterine disease and dairy herd reproductive performance: A review.
      ;
      • Galvão K.N.
      Postpartum uterine diseases in dairy cows.
      ;
      • Vieira-Neto A.
      • Lima F.S.
      • Santos J.E.P.
      • Mingoti R.D.
      • Vasconcellos G.S.
      • Risco C.A.
      • Galvao K.N.
      Vulvovaginal laceration as a risk factor for uterine disease in postpartum dairy cows.
      ). Inconsistency in the definition of metritis across herds and studies may account for the large variation in incidence of metritis (
      • Sannmann I.
      • Arlt S.
      • Heuwieser W.
      A critical evaluation of diagnostic methods used to identify dairy cows with acute post-partum metritis in the current literature.
      ;
      • Espadamala A.
      • Pallarés P.
      • Lago A.
      • Silva-Del-Río N.
      Fresh-cow handling practices and methods for identification of health disorders on 45 dairy farms in California.
      ), and may lead to underestimation of the real effect of metritis in dairy farms (
      • Espadamala A.
      • Pallarés P.
      • Lago A.
      • Silva-Del-Río N.
      Fresh-cow handling practices and methods for identification of health disorders on 45 dairy farms in California.
      ;
      • McCarthy M.M.
      • Overton M.W.
      Short communication: Model for metritis severity predicts that disease misclassification underestimates projected milk production losses.
      ).
      Previous research has shown that cows with metritis had reduced milk yield compared with cows that did not have metritis (
      • Rajala P.J.
      • Gröhn Y.T.
      Effects of dystocia, retained placenta, and metritis on milk yield in dairy cows.
      ;
      • Dubuc J.
      • Duffield T.F.
      • Leslie K.E.
      • Walton J.S.
      • Leblanc S.J.
      Effects of postpartum uterine diseases on milk production and culling in dairy cows.
      ;
      • Pérez-Báez J.
      • Risco C.A.
      • Chebel R.C.
      • Gomes G.C.
      • Greco L.F.
      • Tao S.
      • Thompson I.M.
      • do Amaral B.C.
      • Zenobi M.G.
      • Martinez N.
      • Staples C.R.
      • Dahl G.E.
      • Hernández J.A.
      • Santos J.E.P.
      • Galvão K.N.
      Association of dry matter intake and energy balance prepartum and postpartum with health disorders postpartum: Part I. Calving disorders and metritis.
      ), and that the negative effect of metritis could be greatly underestimated if metritis is inconsistently or poorly diagnosed and recorded (
      • McCarthy M.M.
      • Overton M.W.
      Short communication: Model for metritis severity predicts that disease misclassification underestimates projected milk production losses.
      ). Cows with metritis have also been shown to have reduced resumption of ovarian cyclicity (
      • Santos J.E.P.
      • Bisinotto R.S.
      • Ribeiro E.S.
      • Lima F.S.
      • Greco L.F.
      • Staples C.R.
      • Thatcher W.W.
      Applying nutrition and physiology to improve reproduction in dairy cattle.
      ;
      • Ribeiro E.S.
      • Lima F.S.
      • Greco L.F.
      • Bisinotto R.S.
      • Monteiro A.P.A.
      • Favoreto M.
      • Ayres H.
      • Marsola R.S.
      • Martinez N.
      • Thatcher W.W.
      • Santos J.E.P.
      Prevalence of periparturient diseases and effects on fertility of seasonally calving grazing dairy cows supplemented with concentrates.
      ;
      • Vercouteren M.M.
      • Bittar J.H.
      • Pinedo P.J.
      • Risco C.A.
      • Santos J.E.
      • Vieira-Neto A.
      • Galvão K.N.
      Factors associated with early cyclicity in postpartum dairy cows.
      ), pregnancy per AI (
      • Santos J.E.P.
      • Bisinotto R.S.
      • Ribeiro E.S.
      • Lima F.S.
      • Greco L.F.
      • Staples C.R.
      • Thatcher W.W.
      Applying nutrition and physiology to improve reproduction in dairy cattle.
      ;
      • Mahnani A.
      • Sadeghi-Sefidmazgi A.
      • Cabrera V.E.
      Consequences and economics of metritis in Iranian Holstein dairy farms.
      ;
      • Lima F.S.
      • Vieira-Neto A.
      • Snodgrass J.
      • De Vries A.
      • Santos J.E.
      Economic comparison of systemic antimicrobial therapies for metritis in dairy cows.
      ), and increased calving to conception interval (
      • Giuliodori M.J.
      • Magnasco R.P.
      • Becu-Villalobos D.
      • Lacau-Mengido I.M.
      • Risco C.A.
      • de la Sota R.
      Metritis in dairy cows: Risk factors and reproductive performance.
      ;
      • Vieira-Neto A.
      • Lima F.S.
      • Santos J.E.P.
      • Mingoti R.D.
      • Vasconcellos G.S.
      • Risco C.A.
      • Galvao K.N.
      Vulvovaginal laceration as a risk factor for uterine disease in postpartum dairy cows.
      ) compared with cows without metritis. Furthermore, some studies have observed that cows with metritis were ∼30% more likely to be culled from the herd (
      • Wittrock J.M.
      • Proudfoot K.L.
      • Weary D.M.
      • von Keyserlingk M.A.
      Short communication: Metritis affects milk production and cull rate of Holstein multiparous and primiparous dairy cows differently.
      ), although the effect on culling has been inconsistent (
      • Gröhn Y.T.
      • Eicker S.W.
      • Ducrocq V.
      • Hertl J.A.
      Effect of diseases on the culling of Holstein dairy cows in New York State.
      ;
      • Dubuc J.
      • Duffield T.F.
      • Leslie K.E.
      • Walton J.S.
      • Leblanc S.J.
      Effects of postpartum uterine diseases on milk production and culling in dairy cows.
      ).
      It is intuitive that decreases in milk yield, reproductive performance, and survival in the herd would lead to decreased profitability of cows with metritis compared with healthy cows. Additionally, metritis would incur in treatment costs. Indeed, researchers have estimated the cost of metritis in herds in Michigan (
      • Bartlett P.C.
      • Kirk J.H.
      • Wilke M.A.
      • Kaneene J.B.
      • Mather E.C.
      Metritis complex in Michigan Holstein-Friesian cattle: Incidence, descriptive epidemiology and estimated economic impact.
      ), California (
      • Overton M.
      • Fetrow J.
      Economics of postpartum uterine health.
      ), and Florida (
      • Lima F.S.
      • Vieira-Neto A.
      • Snodgrass J.
      • De Vries A.
      • Santos J.E.
      Economic comparison of systemic antimicrobial therapies for metritis in dairy cows.
      ). These researchers calculated a cost of $106 ($246 inflation adjusted dollars) in Michigan, $358 ($425 inflation adjusted dollars) in California, and between $267 and $406 in Florida. In addition, outside the United States,
      • Mahnani A.
      • Sadeghi-Sefidmazgi A.
      • Cabrera V.E.
      Consequences and economics of metritis in Iranian Holstein dairy farms.
      estimated that a case of metritis in dairy farms in Iran costs from $146 to $176 (mean of $162/case).
      In summary, it is clear that metritis has a negative effect on farm profitability; however, large-scale studies including several herds from different regions of the United States are still lacking. Therefore, the objective of this study was to estimate the cost of metritis in dairy herds using data collected from 16 dairy herds located in 4 different regions of the United States.

      MATERIALS AND METHODS

      Study Design, Herds, and Study Population

      This was a prospective cohort study that collected data from 11,733 Holstein dairy cows (4,102 primiparous and 7,631 multiparous) from 16 different farms located in 4 different regions of the United States [Northeast (2,635 cows in 4 herds), Midwest (5,189 cows in 6 herds), Southeast (1,183 cows in 1 herd), and Southwest (2,726 cows in 5 herds)] that calved from November 2012 to October 2014. Cows were enrolled at parturition and monitored weekly for disease occurrence, reproductive events, and survival. Collection of general data included herd identification, cow identification, lactation number, date of calving, calf sex, and calving-related problems (dystocia, stillbirth, twins, and retained placenta), type of reproductive management (i.e., estrus or timed artificial insemination, TAI), number of inseminations, insemination dates, and days open. Data were collected for up to 305 DIM. These data were collected as part of a USDA-funded grant titled “Translational Genomics for Improved Fertility of Animals” that supported the phenotyping of the cows needed for the grant. The herds included in the study and the study population are described in detail in the study by
      • Pinedo P.
      • Santos J.E.P.
      • Chebel R.C.
      • Galvão K.N.
      • Schuenemann G.M.
      • Bicalho R.C.
      • Gilbert R.O.
      • Rodriguez Zas S.
      • Seabury C.M.
      • Rosa G.
      • Thatcher W.W.
      Early-lactation diseases and fertility in 2 seasons of calving across US dairy herds.
      .

      Sample Estimation

      The current study used a convenience sample of 11,733 Holstein dairy cows from 16 herds; therefore, no a priori sample size calculation was performed. The initial sample size was estimated to identify genomic markers of fertility. For the variable gross profit, approximately 1,400 cows per group would be needed to detect differences of $200 with a standard deviation of $1,888 (
      • Lima F.S.
      • Vieira-Neto A.
      • Snodgrass J.
      • De Vries A.
      • Santos J.E.
      Economic comparison of systemic antimicrobial therapies for metritis in dairy cows.
      ), α of 0.05, and β of 0.2.

      Metritis Diagnosis

      Diagnosis of metritis was performed during weekly visits to the study farms. Metritis was diagnosed by evaluation of the vaginal discharge collected using the Metricheck device (Simcro, Hamilton, New Zealand) at 7 ± 3 DIM. The discharge retrieved from the vagina was scored as 1 = not fetid normal lochia, viscous, clear, red, or brown; 2 = cloudy mucoid discharge with flecks of pus; 3 = not fetid, mucopurulent discharge with <50% pus; 4 = not fetid mucopurulent white, yellow, or reddish-brownish discharge with ≥50% pus; and 5 = fetid, thin, serous, or watery, may have been reddish-brownish, with or without pieces of necrotic tissue present (adapted from
      • Chenault J.R.
      • McAllister J.F.
      • Chester S.T.
      • Dame K.J.
      • Kausche F.M.
      • Robb E.J.
      Efficacy of ceftiofur hydrochloride sterile suspension administered parenterally for the treatment of acute postpartum metritis in dairy cows.
      ). Cows with a vaginal discharge score of 5 were classified as having metritis. It is important to note that there was no communication of the diagnosis with the farm personnel; therefore, we do not have information on whether or when metritis was diagnosed or treated. For this economic analysis, we assumed cows were treated, and we described different treatment scenarios in a separate section below.

      Inclusion and Exclusion Criteria

      Inclusion criteria for cows in this current study included the availability of data related to metritis diagnosis, milk production, reproductive performance, and survival in the herd. The exclusion criteria included cows that had missing data, that were not examined for metritis, or if they were examined for metritis before 4 DIM or after 10 DIM. Therefore, 47 cows were excluded because they were not examined for metritis diagnosis, 60 cows were excluded because they were examined before 4 DIM, and 46 cows were excluded because they were examined after 10 DIM. Therefore, the final data set for this study included a total of 11,581 cows (4,049 primiparous and 7,532 multiparous). Of those, 2,907 were diagnosed with metritis (score of 5) and 8,674 were not diagnosed with metritis (score ≤4), and therefore were classified as without metritis.

      Metritis Treatment Costs

      Metritis treatment cost was the average cost of 4 different treatment scenarios. In the first scenario we assumed that cows with metritis were treated with 6.6 mg of ceftiofur/kg of BW administered subcutaneously (Excede sterile suspension, 200 mg of ceftiofur as ceftiofur crystalline free acid, Zoetis, Kalamazoo, MI) twice with the second dose administered 72 h after the initial dose. The mean BW during the first 14 d postpartum for primiparous (560 kg) and multiparous (680 kg) cows were used to calculate the dose administered. The cost of a 500-mL vial of Excede was set at 523.08 according to current market prices, resulting in a cost of $2.09/mL. In addition to medication, the costs associated with treatment included the supplies for treatment administration and the cost of labor for cow restraint and administration of treatment. Cost of supplies for treatment such as needle and syringe was $0.62 per treatment. Labor cost for performing treatments was priced at $10.8 per hour (), and 4 min per cow, which resulted in $0.72 per treatment. Therefore, the total cost per treatment with Excede for a case of metritis for primiparous and multiparous cows was $80.0 and $96.6, respectively.
      In the second scenario, we assumed that cows with metritis were treated with 2.2 mg of ceftiofur/kg of BW administered intramuscularly (Excenel, 50 mg/mL of ceftiofur, as ceftiofur hydrochloride, Zoetis, Kalamazoo, MI) once daily for 5 d. The cost of a 250-mL vial of Excenel was set at $236.99 according to current market prices, resulting in a cost of $0.96/mL. Body weights and the costs associated with treatment such as the supplies for treatment administration and the cost of the time for cow restraint and administration of treatment were the same as the first scenario. Therefore, the total cost per treatment with Excenel for a case of metritis for primiparous and multiparous cows was $125.0 and $150.3, respectively.
      In the third scenario, we assumed that cows with metritis were treated with 11 mg of ampicillin/kg of BW administered intramuscularly (Polyflex, 250 mg/mL of ampicillin, as ampicillin trihydrate, Boehringer Ingelheim Vetmedica, St. Joseph, MO) once daily for 5 d. The cost of a 100-mL vial of Polyflex was set at $47.11 according to current market prices, resulting in a cost of $0.47/mL. Body weights and the costs associated with treatment such as the supplies for treatment administration and the cost of the time for cow restraint and administration of treatment were the same as the first scenario. Therefore, the cost per treatment was $64.6 and $77.0 for primiparous and multiparous cows, respectively, and to this cost was added the cost of discarded milk multiplied by the duration of milk withholding (i.e., 7 d), resulting in a total cost of ampicillin including discarded milk of $122.8 and $153.3 for primiparous and multiparous cows, respectively. In the fourth scenario, we assumed that cows were treated with ampicillin as explained above, but the milk fed to calves instead of being discarded. Therefore, the value of milk fed to calves during milk withholding was subtracted from the cost of treatment with ampicillin, resulting in a total cost of $83.8 and $114.3 for primiparous and multiparous cows, respectively. Therefore, the average treatment cost for these 4 scenarios was $102.3 and $128.6 for primiparous and multiparous cows, respectively.

      Milk Production and Milk Income Calculation

      Milk production was measured monthly by the DHIA for a total of 10 tests. The total milk yield was calculated averaging the milk production of each test and multiplying this average by the total DIM in the lactation up to 305 DIM. Milk income was calculated by multiplying the total milk yield for each cow by $0.395, which was the mean price per kg of milk sold from 2008 to 2018 (agricultural prices, ).
      Discarded milk was calculated by multiplying the average milk production during the first 14 d, for primiparous (i.e., 21.1 kg) and multiparous (27.6 kg) cows, respectively, by the mean price per kg of milk (i.e., $0.395/kg), which resulted in a total of $8.3 and $10.9, respectively. However, if the milk was fed to calves, then the average milk production during the first 14 d of each parity was multiplied by the current market value of reconstituted milk (i.e., $0.265/kg), which resulted in a total of $5.6 and $7.3 for primiparous and multiparous cows, respectively.

      Estimated Feed Costs

      For calculation of feed costs, we assumed that all lactating cows were fed a TMR with a NEL density of 1.60 Mcal/kg. The milk net energy was assumed at 0.69 Mcal/kg based on milk containing 3.5% fat, 4.8% lactose, and 3.2% protein. Therefore, each kg of marginal DM consumed supported 2.32 kg of milk and the NEL needs for maintenance was calculated as 0.08 Mcal/kg of BW0.75, where BW was calculated by averaging the BW at calving and the BW on the last day of the lactation up to 305 DIM using the following regression equations.
      For primiparous:
      BW (kg) for the first 26 DIM = 584.4 – 3.91 × DIM + 0.081(DIM)2;


      BW (kg) from 27 up to 305 DIM = 520.6 + 0.42 × DIM + 0.00013(DIM)2.


      For multiparous:
      BW (kg) for the first 26 DIM = 701.6 – 3.47 × DIM + 0.055(DIM)2;


      BW (kg) from 27 up to 305 DIM = 647.6 + 0.14 × DIM + 0.00085(DIM)2.


      The regression equations were built using daily BW available for a 305-d lactation from 81 primiparous and 79 multiparous from the University of Florida Dairy Unit in Gainesville. The coefficient of determination (R2) for primiparous cows for the first 26 DIM was 97.1% and from 27 to 305 DIM was 98.9%. For multiparous cows for the first 26 DIM was 92.8% and from 27 to 305 DIM was 98.8%.
      The DMI required to support maintenance was calculated by multiplying each daily maintenance needs in kg of DMI by the number of days in lactation up until the cow died, was sold, or at 305 DIM, whichever came first. The DMI required to support milk production was calculated as the total milk yield in the lactation multiplied by 0.69 and then divided by 1.60. The costs associated with feed for each lactating cow were calculated by multiplying the total DMI (maintenance plus production) by the average cost of a lactating cow TMR from 2010 to 2017 of $0.265/kg of DM ().

      Reproductive Management

      Reproductive management varied among farms but usually included a combination of estrus detection and TAI for first and subsequent AI; therefore, we assumed that all the cows were enrolled in a Presynch-Ovsynch program (
      • Moreira F.
      • Orlandi C.
      • Risco C.A.
      • Mattos R.
      • Lopes F.
      • Thatcher W.W.
      Effects of presynchronization and bovine somatotropin on pregnancy rates to a timed artificial insemination protocol in lactating dairy cows.
      ), with the option of being AI in estrus after the second PGF of the presynchronization. Data for first service were available for 10,487 cows, and 4,973 were AI after estrus detection and 5,514 were TAI. Data for first AI were not available for 1,094 cows because they were culled before performing AI.
      Pregnancy diagnosis was performed using transrectal ultrasonography at 33 ± 3 d after AI and was confirmed by palpation per rectum at 60 ± 3 and 200 ± 3 d after AI. Therefore, cows that were pregnant at the end of the study were assumed to have had 3 pregnancy diagnoses, and cows that did not receive an AI throughout the lactation at the end of the study were assumed to have not been evaluated for pregnancy.

      Reproductive Management Costs

      The cost of reproductive management for cows that were AI after detected estrus for first service included the cost of pre-synchronization, labor cost for hormonal administration, cost of daily heat detection, labor cost of AI, cost of AI, and cost of pregnancy diagnosis. The cost of the reproduction management for cows that were serviced with TAI were calculated assuming that they received the same treatment as a cow that received AI after estrus detection plus the cost of an ovulation synchronization (Ovsynch) program. The cost for Presynch and the Presynch-Ovsynch program were calculated assuming a cost of $2.70/dose for PGF and $2.00/dose for GnRH according to the prices provided by Southwest Florida Milk. The cost of labor for employees that administered the hormones and performed detection of estrus was assumed to be $10.80/h. Assuming that a person can administer 60 injections/h, the cost per injection was $0.18, and to this price we added $0.05 of supplies for injections thus having a total cost of $0.23 for labor and supplies. For detection of estrus, we assumed that a person could check 120 cows/h and that each cow had 24 estrus checks, thus resulting in a value of $0.09 per check. The cost of an AI technician was $120/h, and assuming that a technician can inseminate 30 cows/h, the cost per cow was $4.00/cow (http://www.chrismancattleservices.com/services.html) The AI cost was calculated assuming a semen cost of $15.00/dose, and supplies for each insemination including AI sheath, sleeve, semen applicator, water bath, and chalk was assumed to be $0.50/AI. Therefore, the cost of AI in cows that were inseminated after showing estrus was $27.52/cow and the ones that were bred with TAI was $34.91/cow. In cows that received more than one AI, the cost per extra breeding was calculated averaging the AI cost in cows that were inseminated after showing estrus and the ones that were inseminated with TAI, excluding the cost of Presynch, resulting in $25.82.
      The cost of pregnancy diagnosis was calculated assuming that each pregnant cow had 3 pregnancy checks; therefore, cows that received one AI and became pregnant were assumed to have 3 pregnancy checks. For cows that received one or several AI and did not become pregnant the cost of pregnancy check was multiplied by 0.5 to assume half of the cost of pregnancy diagnosis, for each extra AI. Cows that received several AI and became pregnant were assumed to have 3 pregnancy checks plus the cost of pregnancy check multiplied by 0.5 to assume half of the cost of pregnancy diagnosis, for each extra AI that the cow received and that did not end in pregnancy. In addition, if the cow experienced a pregnancy loss, it was assumed to have one extra pregnancy check. The veterinarian cost for pregnancy diagnosis was $140.00/h; assuming that a veterinarian can do 30 pregnancy diagnoses/h by ultrasonography, this resulted in a cost per pregnancy check of $4.67/cow.

      Cow Sales, Replacement Costs, and Residual Cow Value

      Survival in the herd was observed until 305 DIM. Cows that survived until 305 DIM were censored at that point. Cows that were not pregnant by 305 DIM were considered to have been sold at 305 DIM. Cows that died or were sold were replaced by a first-lactation cow. Cow sale value was calculated using the formula
      cow sale value = $1.65/kg × BW (kg) at time (DIM) of sale,


      where $1.65/kg was the average beef price in the United States between 2008 and 2017 () and BW at the time of sale for primiparous and multiparous was calculated using the regression formulas presented above.
      A residual cow value was calculated for cows that were pregnant at the end of a 305-d lactation using the formula:
      residual cow value = replacement cost – [lactation no. × (replacement cost – average salvage value)/4],


      where replacement cost was the average price ($1,576.2) for a replacement heifer from 2008 to 2018 (
      • USDA NASS
      2008-2018. Replacement Price.
      ); lactation no. was 1, 2, 3, or 4 (for cows with lactation no. ≥ 4); and average salvage value was calculated as follows:
      average salvage value = [% of primiparous cows sold × average BW (kg) of primiparous cows × $1.65/kg] + [% of multiparous cows sold × average BW (kg) of multiparous cows × $1.65/kg],


      where % of primiparous cows sold was 25.6%, % of multiparous cows sold was 74.3%, average BW of primiparous cows was 591 kg, and the average BW of multiparous cows was 697 kg.

      Gross Profit Calculation

      The gross profit for each cow was calculated using an Excel (Microsoft Corporation, Redmond, WA) spreadsheet and included the income and expenses during their lactation up to 305 d. Each cow generated income from the sale of milk and from their salvage when they were culled. Additional income for each cow was generated by the residual cow value at the end of the 305-d lactation. Expenses for each cow included those incurred with replacement cost, costs associated with reproductive management, therapeutic costs of metritis, and feed cost. Therefore, the formula for gross profit calculation was as follows:
      gross profit = (milk income + salvage value + cow value) – (treatment cost + reproduction cost + DMI cost + replacement cost).


      Stochastic Analysis

      A stochastic Monte Carlo simulation model was developed using @Risk version 7.6 (Palisade Corporation, Ithaca, NY) to estimate differences in cash flow between cows with and without metritis under varying market conditions for the most relevant input. Outcomes of interest were the distribution of differences in cash flow between cows with metritis and without metritis and the contribution of individual input costs to the total variation in cash flow differences.
      Milk production, DMI, percentage of cows sold by 305 DIM, percentage of cows that were culled by 305 DIM, percentage of cows with reproduction cost, percentage of cows not culled, and percentage of cows with metritis were used as fixed inputs (Supplemental Table S1; https://doi.org/10.3168/jds.2020-19125). Stochasticity in every iteration of the simulation was introduced for milk price, replacement cost, salvage value, feed cost, cow value, metritis treatment cost, and reproductive cost (Supplemental Table S2; https://doi.org/10.3168/jds.2020-19125). Monthly milk price, replacement cost, salvage value, feed cost, cow value, metritis treatment cost, and reproductive cost were collected and calculated as explained previously. Distributions used in the simulations for variables were fitted using the BatchFit function of @Risk. This function selects the best-fitting distribution based on the lowest value for the Akaike information criteria. Specifically, milk price was fitted with a log-logistic distribution, replacement cost with a uniform distribution, and salvage value with a Kumaraswamy distribution. Feed cost was fitted with a Pareto distribution, cow value with a triangular distribution, and treatment cost with a uniform distribution. Reproductive cost was fitted with a Weibull distribution (Supplemental Table S2; https://doi.org/10.3168/jds.2020-19125).
      Simulations were run and recorded for 10,000 iterations. For each iteration, the difference in cash flow between cows with metritis and cows without metritis was calculated as
      difference in cash flow ($) = cash flow metritis − cash flow no metritis,


      where cash flow = (milk production × milk price + % cows sold × salvage value + % cows not culled × cow value) – (DMI × feed cost + % of cows culled × replacement cost + % of cows with reproductive cost × reproduction cost + metritis treatment cost).

      Statistical Analysis

      Continuous outcomes such as milk production (kg/cow), milk sales ($/cow), cow sales ($/cow), replacement costs ($/cow), cost of reproduction ($/cow), breeding costs ($/cow), pregnancy diagnosis costs ($/cow), feed costs ($/cow), and gross profit ($/cow) for cows with and without metritis were compared using mixed effect models using the MIXED procedure of SAS (SAS Institute Inc., Cary, NC). Distribution of gross profit was assessed using the Minitab software version 17 (Minitab LLC, State College, PA), and because gross profit was skewed to the left (Supplemental Figure S1; https://doi.org/10.3168/jds.2020-19125), we also analyzed it nonparametrically using the Kruskal–Wallis test in Minitab, and report the median gross profit in addition to the mean gross profit. Analysis of non-normally distributed data using parametric methods can lead to type I or type II errors, particularly with a small sample size, which is less of a concern herein (
      • Vickers A.J.
      Parametric versus non-parametric statistics in the analysis of randomized trials with non-normally distributed data.
      ). Dichotomous outcomes such as pregnancy by 305 DIM, sold, died, and overall culling by 305 DIM were analyzed using the GLIMMIX procedure of SAS with a logit link. All models included the fixed effects of metritis (yes or no), parity (primiparous or multiparous), the interaction between metritis and parity, and farm as the random effect. A stepwise backward elimination was performed and the interaction between metritis and parity, and then parity, were removed from the model if P > 0.10. Variables were considered significant when P ≤ 0.05, and tendency when variables were between P > 0.05 and P ≤ 0.10. The cost of metritis was the difference between the gross profit in cows with metritis and cows without metritis. Distribution of residuals and homogeneity of variance were evaluated for each variable analyzed. The Kenward-Roger approximation method was used to calculate the denominator degrees of freedom for the F-tests in the mixed models and the containment method was used to calculate the denominator degrees of freedom in the generalized linear models. The LSM and SEM were used to report the results.

      RESULTS

      The proportion of cows with metritis in the whole population was 25.3% (2,971/11,733), and in the selected population was 25.1% (2,907/11,581). The inputs used to estimate the cost of metritis are presented in Table 1.
      Table 1Variables and measurements to estimate the cost of metritis
      Input variableMeasure
      Proportion of cows with metritis, %25.1
      Avg BW at metritis diagnosis in primiparous cows, kg560.0
      Avg BW at metritis diagnosis in multiparous cows, kg680.0
      Milk price,
      Source: agricultural prices (USDA NASS, 2017a).
      $/kg
      0.395
      Cost of leaving the herd
       Beef price,
      Source: agricultural prices (USDA NASS, 2017a).
      $/kg of BW
      1.65
       Replacement cost, $/cow1,576.2
      Value of a cow at the end of 1st lactation, $/cow1,458.2
      Value of a cow at the end of 2nd lactation, $/cow1,340.2
      Value of a cow at the end of 3rd lactation, $/cow1,222.2
      Value of a cow at the end of 4th or greater lactation, $/cow1,104.1
      Labor, $/h10.8
      Treatment cost
       Labor and supplies, $/dose1.34
       Metritis treatment in primiparous cows,
      Average (Avg) treatment cost for primiparous cows.
      $/cow
      102.9
       Metritis treatment in multiparous cows,
      Average treatment cost for multiparous cows.
      $/cow
      128.6
      Reproduction cost
       Labor estrus detection (120 detections/h), $/estrus0.09
       Technician for AI (30 AI/h), $/AI4.0
       Semen straw, $/straw15.0
       AI supplies, $/AI0.5
       Cost of PGF, $/unit2.7
       Cost of GnRH, $/unit2.0
       Cost of labor and supplies for hormones injections0.23
      Feed cost
       Diet net energy, Mcal/kg1.60
       Milk net energy, Mcal/kg0.69
       Marginal feed, kg of DM/kg of milk0.43
       Feed cost,
      Source: Economic Research Service (USDA ERS, 2018).
      $/kg
      0.265
      1 Source: agricultural prices ().
      2 Average (Avg) treatment cost for primiparous cows.
      3 Average treatment cost for multiparous cows.
      4 Source: Economic Research Service ().

      Milk Yield and Milk Sales by 305 DIM

      Cows with metritis had lesser milk yield (9,463.0 ± 318.7 vs. 10,277.0 ± 313.8 kg/cow; P < 0.01), which resulted in lesser income from milk sales (3,737.9 ± 125.9 vs. 4,059.4 ± 124.0 $/cow; P < 0.01) than cows without metritis (Table 2).
      Table 2Productive, reproductive, and economic outcomes according to disease status
      ItemMetritis ± SENo metritis ± SEDiff
      Diff = difference between cows that developed metritis and cows that did not develop metritis.
      P-value
      Milk by 305 DIM, kg9,463.0 ± 318.710,277.0 ± 313.8−813.9<0.01
      Pregnant by 305 DIM, %69.2 ± 1.279.2 ± 1.1−10.0<0.01
      Culled by 305 DIM, %35.9 ± 1.326.5 ± 1.19.4<0.01
      Sold, %31.3 ± 1.523.6 ± 1.37.7<0.01
      Died, %4.6 ± 0.62.9 ± 0.51.7<0.01
      DMI, kg5,769.7 ± 147.36,227.1 ± 143.9−457.5<0.01
      Milk sales by 305 DIM, $/cow3,737.9 ± 125.94,059.4 ± 124.0−321.5<0.01
      Cow sales, $/cow338.4 ± 15.7256.6 ± 13.981.8<0.01
      Residual cow value879.4 ± 17.41,004.8 ± 14.9−125.4<0.01
      Feeding costs by 305 DIM, $/cow1,529.0 ± 39.01,650.2 ± 38.0−121.3<0.01
      Replacement costs, $/cow566.1 ± 20.9418.0 ± 17.9148.1<0.01
      Cost of reproduction, $/cow80.4 ± 2.281.0 ± 2.0−0.60.61
      Treatment costs, $/cow117.9 ± 0.10.0 ± 0.0117.9<0.01
      Gross profit, $/cow2,662.2 ± 85.43,172.9 ± 81.9−510.7<0.01
      1 Diff = difference between cows that developed metritis and cows that did not develop metritis.

      Proportion of Cows Pregnant, Culling, and Cow Sales by 305 DIM

      Proportion of cows pregnant by 305 DIM was lesser (69.2 ± 1.2 vs. 79.2 ± 1.1%; P < 0.01) and proportion of cows culled by 305 DIM was greater for cows with metritis than cows without metritis (35.9 ± 1.3 vs. 26.5 ± 1.1%; P < 0.01). Cows that left the herd by 305 DIM could have been sold or dead. Cows with metritis had greater proportion of cows sold by 305 DIM (31.3 ± 1.5 vs. 23.6 ± 1.3%; P < 0.01), which resulted in greater income from cow sales (338.4 ± 15.7 vs. 256.6 ± 13.9 $/cow) than cows without metritis (Table 2). Proportion of cows dead by 305 DIM was also greater for cows with metritis than for cows without metritis (4.6 ± 0.6 vs. 2.9 ± 0.5%; P < 0.01).

      Reproduction and Cow Replacement Costs

      The cost of reproduction did not differ between cows that with metritis and cows without metritis (80.4 ± 2.2 vs. 81.0 ± 2.0 $/cow; P = 0.61), but the cost of replacement was greater (P < 0.01) in cows with metritis compared with cows without metritis (566.1 ± 20.9 vs. 418.0 ± 17.9 $/cow; Table 2).

      DMI and Feed Costs

      Cows with metritis had lesser DMI (5,769.7 ± 147.3 vs. 6,227.1 ± 143.9 kg/cow; P < 0.01), which resulted in lesser feeding costs (1,529.0 ± 39.0 vs. 1,650.2 ± 38.1 $/cow; P < 0.01) than cows without metritis.

      Treatment Costs

      The cost of metritis treatment averaged $117.9 ± 0.1.

      Gross Profit

      Cows with metritis had lesser overall gross profit than cows without metritis. The mean gross profit for cows with metritis was lesser than for cows without metritis (2,662.2 ± 85.4 vs. 3,172.9 ± 81.9 $/cow; P < 0.01; Table 2), and the median gross profit was also lesser for cows with metritis than for cows without metritis (3,257.2 vs. 3,655.5 $/cow; P < 0.01). The difference in mean gross profit for cows with metritis and cows without metritis was −$510.7 per cow but ranged from −$156.1 to −$947.6 depending on the herd (Supplemental Table S3 and Figure S2; https://doi.org/10.3168/jds.2020-19125), and the difference between medians was $398.4.

      Stochastic Analysis

      The mean difference in cash flow for cows with metritis and without metritis in all scenarios simulated by the @Risk program was −$512.8, with 95% of the scenarios ranging from −$884.1 to −$240.3 (Figure 1A). The minimum difference in cash flow for cows with metritis and without metritis was −$1,000 and the maximum was −$252. Milk price had the greatest effect on metritis mean cost and accounted for 58.7% of the variation (Figure 1B). Treatment cost, replacement cost, feeding cost, cow value, and salvage value contributed 19.0, 11.7, 7.2, 2.0, and 1.4% of the variation, respectively (Figure 1B). Spearman correlation coefficients showed that milk price had the highest correlation (ρ = −0.73), followed by treatment cost (ρ = −0.45), replacement cost (ρ = −0.35), feed price (ρ = 0.21), cow value (ρ = −0.12), and salvage value (ρ = 0.12). Mean change across percentile range of the top 5 input variables included in the stochastic simulation model is shown in Figure 1C.
      Figure thumbnail gr1
      Figure 1(A) Relative frequency distribution for differences in cash flow for cows with metritis and with no metritis for 10,000 iterations of stochastic simulation. Differences in cash flow were calculated by subtracting cash flow of cows with metritis from cows with no metritis. (B) Tornado graph showing the contribution to the variance of the variables included in the stochastic simulation model. (C) Mean change across percentile range of the top 5 input variables included in the stochastic simulation model.

      DISCUSSION

      The objective of this study was to estimate the cost of metritis in dairy herds from the United States. We estimated that a case of metritis costs approximately $512, ranging from $156 to $948 depending on the farm, and from $240 to $884 depending on the simulation scenario. Cows with metritis had lesser milk yield and proportion of pregnant cows by 305 DIM, and greater culling, which led to decreased income from milk sales and higher replacement costs. Nonetheless, milk price was the greatest contributor to the variation in metritis cost.
      The cost of metritis presented in this study is higher than previously reported.
      • Bartlett P.C.
      • Kirk J.H.
      • Wilke M.A.
      • Kaneene J.B.
      • Mather E.C.
      Metritis complex in Michigan Holstein-Friesian cattle: Incidence, descriptive epidemiology and estimated economic impact.
      estimated that the total cost per metritis case was $106 ($246 inflation adjusted) in dairy herds from Michigan state. The main difference between studies is that in
      • Bartlett P.C.
      • Kirk J.H.
      • Wilke M.A.
      • Kaneene J.B.
      • Mather E.C.
      Metritis complex in Michigan Holstein-Friesian cattle: Incidence, descriptive epidemiology and estimated economic impact.
      milk yield loss was not included as an input for the calculation of metritis cost because the difference in milk yield was not significant between cows that developed metritis and cows that did not develop metritis; instead, they calculated the cost of milk that was withheld after antibiotic treatment, which was $23.85 ($53.68 inflation adjusted) per case of metritis. Conversely, in our study, the difference in milk sales (−$322/cow) resultant of the difference in milk yield (814 kg/lactation) was the largest contributor for the difference in gross profit between cows with metritis and cows without metritis. Additionally, in
      • Bartlett P.C.
      • Kirk J.H.
      • Wilke M.A.
      • Kaneene J.B.
      • Mather E.C.
      Metritis complex in Michigan Holstein-Friesian cattle: Incidence, descriptive epidemiology and estimated economic impact.
      , treatment cost was also very low [2.74 ($6.39 inflation adjusted)] compared with the costs calculated herein, which averaged $118.
      • Overton M.
      • Fetrow J.
      Economics of postpartum uterine health.
      estimated that a case of metritis had an average cost of $358 ($425 inflation adjusted), which is lower than what we reported herein. This difference may be because in the study by
      • Overton M.
      • Fetrow J.
      Economics of postpartum uterine health.
      the milk loss (∼290 kg/lactation) and milk price [$0.29/kg ($0.34 inflation adjusted)] were lower than what was reported herein, which led to lower losses from milk sales [$83 ($100 inflation adjusted)].
      • McCarthy M.M.
      • Overton M.W.
      Short communication: Model for metritis severity predicts that disease misclassification underestimates projected milk production losses.
      reported milk yield losses of 384 kg/lactation for cases of mild metritis and 847 kg/lactation for severe metritis in one herd in Michigan.
      • Lima F.S.
      • Vieira-Neto A.
      • Snodgrass J.
      • De Vries A.
      • Santos J.E.
      Economic comparison of systemic antimicrobial therapies for metritis in dairy cows.
      estimated that metritis cost ranged from $267 to $410, depending on the antibiotic treatment administered and whether waste milk was discarded or fed to the calves. Among the differences between studies that may contribute to differences in the cost of metritis,
      • Lima F.S.
      • Vieira-Neto A.
      • Snodgrass J.
      • De Vries A.
      • Santos J.E.
      Economic comparison of systemic antimicrobial therapies for metritis in dairy cows.
      reported milk losses ranging from 151 to 706 kg/lactation, which resulted in differences in milk sales of −$138 to −$314/cow. Additionally,
      • Lima F.S.
      • Vieira-Neto A.
      • Snodgrass J.
      • De Vries A.
      • Santos J.E.
      Economic comparison of systemic antimicrobial therapies for metritis in dairy cows.
      only considered culling that occurred before 300 DIM, whereas we classified cows that were not pregnant at the end of the lactation as culled, therefore decreasing residual cow value and increasing replacement cost for cows with metritis. It is also important to mention that both
      • Overton M.
      • Fetrow J.
      Economics of postpartum uterine health.
      and
      • Lima F.S.
      • Vieira-Neto A.
      • Snodgrass J.
      • De Vries A.
      • Santos J.E.
      Economic comparison of systemic antimicrobial therapies for metritis in dairy cows.
      estimated the cost of metritis in only one herd. We observed great variability in milk losses and the cost of metritis across herds. Milk losses ranged from 325 to 2,203 kg/lactation, whereas profit losses ranged from $215 to $973/cow (Supplemental Table S3; https://doi.org/10.3168/jds.2020-19125). Using a herd simulation model,
      • Liang D.
      • Arnold L.M.
      • Stowe C.J.
      • Harmon R.J.
      • Bewley J.M.
      Estimating US dairy clinical disease costs with a stochastic simulation model.
      estimated that a case of metritis had an average cost of $172 in primiparous and $263 in multiparous cows. This is also lower than what was reported herein, mostly because of much lower estimated losses from milk sales, which were only $3.3 per case in primiparous and $9.2 per case in multiparous cows. Therefore, we believe our study provides a good representation of the current expected average cost of metritis in dairy herds in the United States. Additionally, the variation in milk yield and profit losses due to metritis among herds indicate that farm management practices may influence productive and economic outcomes in cows with metritis. For this study, the team of investigators performed the diagnosis of metritis; however, there was no communication of the diagnosis with the farm personnel. Therefore, whether metritis was diagnosed or not, timing of metritis diagnosis and treatment by the farm personnel when metritis was diagnosed and treated, and treatment regimen may be contributing factors to the differences observed among herds, and need further investigation.
      Studies outside the United States have also estimated the cost of metritis.
      • Drillich M.
      • Beetz O.
      • Pfützner A.
      • Sabin M.
      • Sabin H.J.
      • Kutzer P.
      • Nattermann H.
      • Heuwieser W.
      Evaluation of a systemic antibiotic treatment of toxic puerperal metritis in dairy cows.
      estimated that mean cost of metritis in herds in Germany ranged from $291 ($420 inflation adjusted) to $362 ($523 inflation adjusted) depending on the treatment used, which is very similar to our results. On the other hand,
      • Mahnani A.
      • Sadeghi-Sefidmazgi A.
      • Cabrera V.E.
      Consequences and economics of metritis in Iranian Holstein dairy farms.
      estimated the mean cost of metritis in herds in Iran as $162 per case, which is much lower than what is reported herein. One major difference is that in the study by
      • Mahnani A.
      • Sadeghi-Sefidmazgi A.
      • Cabrera V.E.
      Consequences and economics of metritis in Iranian Holstein dairy farms.
      the milk yield loss for cows with metritis used in the estimation was only 196.6 kg, and the milk price was $0.32/kg, which resulted in a loss of $63/cow. Treatment cost in the study by
      • Mahnani A.
      • Sadeghi-Sefidmazgi A.
      • Cabrera V.E.
      Consequences and economics of metritis in Iranian Holstein dairy farms.
      was only $30, accounting for veterinary services and drug costs, comparted with $118 in our study. Finally,
      • Mahnani A.
      • Sadeghi-Sefidmazgi A.
      • Cabrera V.E.
      Consequences and economics of metritis in Iranian Holstein dairy farms.
      reports that the main contributor to the cost of metritis was decreased fertility (35 to 47% of total costs), which was accounted as a cost for each additional day open ($4/d), whereas in our study decreased fertility led to increased culling, increased replacement costs, and decreased residual cow value.
      One important point that is usually not addressed in an economic analysis is the effect of selection bias due to loss to follow-up (
      • Østergaard S.
      • Gröhn Y.T.
      Effect of diseases on test day milk yield and body weight of dairy cows from Danish research herds.
      ;
      • de Oliveira E.B.
      • Cunha F.
      • Daetz R.
      • Figueiredo C.C.
      • Chebel R.C.
      • Santos J.E.
      • Risco C.A.
      • Jeong K.C.
      • Machado V.S.
      • Galvão K.N.
      Using chitosan microparticles to treat metritis in lactating dairy cows.
      ). Because cows that are sold or die cannot produce milk or get pregnant, it becomes difficult to estimate the true effect of a disease on those variables and consequently to estimate the true cost of a disease. When trying to estimate the effect of a disease on milk yield and fertility, multivariable regression models are available and have been used to adjust for loss to follow-up due to culling (
      • Østergaard S.
      • Gröhn Y.T.
      Effect of diseases on test day milk yield and body weight of dairy cows from Danish research herds.
      ;
      • de Oliveira E.B.
      • Cunha F.
      • Daetz R.
      • Figueiredo C.C.
      • Chebel R.C.
      • Santos J.E.
      • Risco C.A.
      • Jeong K.C.
      • Machado V.S.
      • Galvão K.N.
      Using chitosan microparticles to treat metritis in lactating dairy cows.
      ). Nonetheless, in an economic analysis, the real data collected from the cows are used and not estimated, and in general, an early culling leads to greater losses because of decreased income from milk sales and increased replacement costs. Herein, we observed that cows with metritis had increased culling up to 100 DIM (Supplemental Figure S3; https://doi.org/10.3168/jds.2020-19125); therefore, we compared the gross profit of cows culled before and after 100 DIM and saw that the mean gross profit in cows culled before 100 DIM was −$555, whereas the mean gross profit in cows culled after 100 DIM was $3,348 (Supplemental Figure S4; https://doi.org/10.3168/jds.2020-19125). Therefore, because of greater early culling in cows with metritis and because of greater gross profit losses in cows with early culling, the loss in profit in cows with metritis is less likely to be underestimated.
      The stochastic simulation with 10,000 different pricing scenarios aligned well with the observed results from the herds studied. The simulation showed that milk price had the greatest contribution to the variation in metritis cost, and losses increased as milk prices increased. This finding is similar to the study by
      • Lima F.S.
      • Vieira-Neto A.
      • Snodgrass J.
      • De Vries A.
      • Santos J.E.
      Economic comparison of systemic antimicrobial therapies for metritis in dairy cows.
      , which showed that milk price accounted for 40 to 57% of the cost of metritis, and higher milk prices resulted in higher profit losses from metritis. Likewise,
      • Lima F.S.
      • Vieira-Neto A.
      • Snodgrass J.
      • De Vries A.
      • Santos J.E.
      Economic comparison of systemic antimicrobial therapies for metritis in dairy cows.
      showed that treatment cost was the second most important contributor to metritis cost, accounting for 17 to 25% of the cost of metritis, and that higher treatment cost resulted in greater metritis cost.

      CONCLUSIONS

      Cows with metritis had decreased milk yield, decreased proportion of cows pregnant, and increased proportion of cows culled at the end of a 305-d lactation. This led to decreased milk sales, feeding costs, and residual cow value, and increased replacement costs and income with cow sales, which led a decrease in profit of approximately $512. Both farm level and stochastic simulation showed great variability in metritis cost, which ranged from $156 to $948 depending on the farm or simulation scenario. The most important variable that contributed to the total variance was milk price followed by treatment cost, replacement cost, and feeding cost. Altogether these 4 variables explained ∼97% of the total variation in cash flow differences. In summary, metritis caused large economic losses to dairy herds by decreasing milk production, reproduction, and survival in the herd.

      ACKNOWLEDGMENTS

      The authors acknowledge the USDA, Washington, DC (National Institute of Food and Agriculture–Agriculture and Food Research Initiative Translational Genomics for Improved Fertility of Animals grant #2013-68004) for financial support to perform the phenotyping of the study animals. We thank the participant dairy farms that allowed all the continuous weekly monitoring of cows. The authors declare that they have no competing interests.

      Supplementary Material

      REFERENCES

        • Bartlett P.C.
        • Kirk J.H.
        • Wilke M.A.
        • Kaneene J.B.
        • Mather E.C.
        Metritis complex in Michigan Holstein-Friesian cattle: Incidence, descriptive epidemiology and estimated economic impact.
        Prev. Vet. Med. 1986; 4: 235-248
        • Chenault J.R.
        • McAllister J.F.
        • Chester S.T.
        • Dame K.J.
        • Kausche F.M.
        • Robb E.J.
        Efficacy of ceftiofur hydrochloride sterile suspension administered parenterally for the treatment of acute postpartum metritis in dairy cows.
        J. Am. Vet. Med. Assoc. 2004; 224 (15154734): 1634-1639
        • de Oliveira E.B.
        • Cunha F.
        • Daetz R.
        • Figueiredo C.C.
        • Chebel R.C.
        • Santos J.E.
        • Risco C.A.
        • Jeong K.C.
        • Machado V.S.
        • Galvão K.N.
        Using chitosan microparticles to treat metritis in lactating dairy cows.
        J. Dairy Sci. 2020; 103 (32505402): 7377-7391
        • Drillich M.
        • Beetz O.
        • Pfützner A.
        • Sabin M.
        • Sabin H.J.
        • Kutzer P.
        • Nattermann H.
        • Heuwieser W.
        Evaluation of a systemic antibiotic treatment of toxic puerperal metritis in dairy cows.
        J. Dairy Sci. 2001; 84 (11573780): 2010-2017
        • Dubuc J.
        • Duffield T.F.
        • Leslie K.E.
        • Walton J.S.
        • Leblanc S.J.
        Effects of postpartum uterine diseases on milk production and culling in dairy cows.
        J. Dairy Sci. 2011; 94 (21338799): 1339-1346
        • Espadamala A.
        • Pallarés P.
        • Lago A.
        • Silva-Del-Río N.
        Fresh-cow handling practices and methods for identification of health disorders on 45 dairy farms in California.
        J. Dairy Sci. 2016; 99 (27592441): 9319-9333
        • Galvão K.N.
        Postpartum uterine diseases in dairy cows.
        Anim. Reprod. 2012; 9: 290-296
        • Giuliodori M.J.
        • Magnasco R.P.
        • Becu-Villalobos D.
        • Lacau-Mengido I.M.
        • Risco C.A.
        • de la Sota R.
        Metritis in dairy cows: Risk factors and reproductive performance.
        J. Dairy Sci. 2013; 96 (23548288): 3621-3631
        • Gröhn Y.T.
        • Eicker S.W.
        • Ducrocq V.
        • Hertl J.A.
        Effect of diseases on the culling of Holstein dairy cows in New York State.
        J. Dairy Sci. 1998; 81 (9594385): 966-978
        • LeBlanc S.J.
        Postpartum uterine disease and dairy herd reproductive performance: A review.
        Vet. J. 2008; 176 (18328749): 102-114
        • Liang D.
        • Arnold L.M.
        • Stowe C.J.
        • Harmon R.J.
        • Bewley J.M.
        Estimating US dairy clinical disease costs with a stochastic simulation model.
        J. Dairy Sci. 2017; 100 (28012631): 1472-1486
        • Lima F.S.
        • Vieira-Neto A.
        • Snodgrass J.
        • De Vries A.
        • Santos J.E.
        Economic comparison of systemic antimicrobial therapies for metritis in dairy cows.
        J. Dairy Sci. 2019; 102 (31178192): 7345-7358
        • Mahnani A.
        • Sadeghi-Sefidmazgi A.
        • Cabrera V.E.
        Consequences and economics of metritis in Iranian Holstein dairy farms.
        J. Dairy Sci. 2015; 98 (26117350): 6048-6057
        • McCarthy M.M.
        • Overton M.W.
        Short communication: Model for metritis severity predicts that disease misclassification underestimates projected milk production losses.
        J. Dairy Sci. 2018; 101 (29550133): 5434-5438
        • Moreira F.
        • Orlandi C.
        • Risco C.A.
        • Mattos R.
        • Lopes F.
        • Thatcher W.W.
        Effects of presynchronization and bovine somatotropin on pregnancy rates to a timed artificial insemination protocol in lactating dairy cows.
        J. Dairy Sci. 2001; 84 (11467815): 1646-1659
        • Østergaard S.
        • Gröhn Y.T.
        Effect of diseases on test day milk yield and body weight of dairy cows from Danish research herds.
        J. Dairy Sci. 1999; 82 (10386305): 1188-1201
        • Overton M.
        • Fetrow J.
        Economics of postpartum uterine health.
        in: Proceedings of the Dairy Cattle Reproduction Council Convention, Omaha, NE. DCRC, Hartland, WI2008: 39-43
        • Pérez-Báez J.
        • Risco C.A.
        • Chebel R.C.
        • Gomes G.C.
        • Greco L.F.
        • Tao S.
        • Thompson I.M.
        • do Amaral B.C.
        • Zenobi M.G.
        • Martinez N.
        • Staples C.R.
        • Dahl G.E.
        • Hernández J.A.
        • Santos J.E.P.
        • Galvão K.N.
        Association of dry matter intake and energy balance prepartum and postpartum with health disorders postpartum: Part I. Calving disorders and metritis.
        J. Dairy Sci. 2019; 102 (31326177): 9138-9150
        • Pinedo P.
        • Santos J.E.P.
        • Chebel R.C.
        • Galvão K.N.
        • Schuenemann G.M.
        • Bicalho R.C.
        • Gilbert R.O.
        • Rodriguez Zas S.
        • Seabury C.M.
        • Rosa G.
        • Thatcher W.W.
        Early-lactation diseases and fertility in 2 seasons of calving across US dairy herds.
        J. Dairy Sci. 2020; 103: 10560-10576
        • Rajala P.J.
        • Gröhn Y.T.
        Effects of dystocia, retained placenta, and metritis on milk yield in dairy cows.
        J. Dairy Sci. 1998; 81 (9891264): 3172-3181
        • Ribeiro E.S.
        • Lima F.S.
        • Greco L.F.
        • Bisinotto R.S.
        • Monteiro A.P.A.
        • Favoreto M.
        • Ayres H.
        • Marsola R.S.
        • Martinez N.
        • Thatcher W.W.
        • Santos J.E.P.
        Prevalence of periparturient diseases and effects on fertility of seasonally calving grazing dairy cows supplemented with concentrates.
        J. Dairy Sci. 2013; 96 (23831093): 5682-5697
        • Sannmann I.
        • Arlt S.
        • Heuwieser W.
        A critical evaluation of diagnostic methods used to identify dairy cows with acute post-partum metritis in the current literature.
        J. Dairy Res. 2012; 79 (22963732): 436-444
        • Santos J.E.P.
        • Bisinotto R.S.
        • Ribeiro E.S.
        • Lima F.S.
        • Greco L.F.
        • Staples C.R.
        • Thatcher W.W.
        Applying nutrition and physiology to improve reproduction in dairy cattle.
        Soc. Reprod. Fertil. Suppl. 2010; 7 (21755686): 385-403
        • USDA ERS
        Feed cost. National feed cost production.
        • USDA NASS
        Farm Labor.
        • USDA NASS
        Milk Price.
        • USDA NASS
        Beef Price.
        • USDA NASS
        2008-2018. Replacement Price.
        • Vercouteren M.M.
        • Bittar J.H.
        • Pinedo P.J.
        • Risco C.A.
        • Santos J.E.
        • Vieira-Neto A.
        • Galvão K.N.
        Factors associated with early cyclicity in postpartum dairy cows.
        J. Dairy Sci. 2015; 98 (25465552): 229-239
        • Vickers A.J.
        Parametric versus non-parametric statistics in the analysis of randomized trials with non-normally distributed data.
        BMC Med. Res. Methodol. 2005; 5 (16269081): 35
        • Vieira-Neto A.
        • Lima F.S.
        • Santos J.E.P.
        • Mingoti R.D.
        • Vasconcellos G.S.
        • Risco C.A.
        • Galvao K.N.
        Vulvovaginal laceration as a risk factor for uterine disease in postpartum dairy cows.
        J. Dairy Sci. 2016; 99 (27016827): 4629-4637
        • Wittrock J.M.
        • Proudfoot K.L.
        • Weary D.M.
        • von Keyserlingk M.A.
        Short communication: Metritis affects milk production and cull rate of Holstein multiparous and primiparous dairy cows differently.
        J. Dairy Sci. 2011; 94 (21524531): 2408-2412