Advertisement

Relationships between management factors in dairy production systems and mental health of farm managers in Japan

Open AccessPublished:November 08, 2021DOI:https://doi.org/10.3168/jds.2021-20666

      ABSTRACT

      To facilitate sustainable dairy farming, it is essential to assess and support the mental health of dairy farm workers, which is affected more than that of workers in other industries, as indicated by the relatively few studies to date. In addition, the limited investigations on mental health in dairy workers minimize the opportunities to suggest practical approaches of improvement of their mental health. Therefore, further data acquisition and analysis is required. In the present study, we undertook quantitative surveys on 17 management factors and administered a mental health questionnaire to 81 dairy farm managers (80 male, 1 female) in Hokkaido, northern Japan. The management factors were categorized into 3 groups: production input, production output, and facility indicator; mental health was evaluated based on the Center for Epidemiologic Studies Depression Scale (CES-D). Principal component analysis assigned the factors into 2 groups: intensiveness factors of dairy production systems (PC1: livestock care cost, fat- and protein-corrected milk, stocking density, medical consultation fee per unit time per animal unit, nonfamily wages, fertilizer and pesticide expenses, and net agricultural income ratio) and basic dairy management factors (PC2: net agricultural income ratio, quantity of concentrate feed, and milk quality variable). The depression symptoms of dairy farm managers were not significantly associated with PC1 and milking methods; however, they were significantly negatively associated with PC2, which integrated 3 management factors, including factors related to finances, feeding, and milk quality. According to the findings of the present study, the efforts needed for stable economic farm management, adequate feed supply, and milk quality maintenance may increase the depression levels of dairy farm managers and negatively affect their mental health. These findings could be the basis for future studies on the relationship between the mental health of farm managers and sustainable dairy farm management and production.

      Key words

      INTRODUCTION

      In recent years, the Japanese dairy farming industry has experienced several changes. The signing of the Comprehensive and Progressive Agreement for Trans-Pacific Partnership by 11 countries, including Japan, on December 30, 2018, which excludes tariffs on agricultural products, has increased the import of agricultural products. Moreover, over the years, Japan's food self-sufficiency rate has been on a downward trend, with approximately 37% based on calorie and 66% based on production value in 2018, compared with 43% and 74%, respectively, in 1995 (
      • MAFF
      Prefectural Food Self-Sufficiency Rate, 2018.
      ).
      The aging of farm workers is also a severe problem influencing agriculture in Japan; the average age of individuals mainly engaged in farming was 67.8 years (
      • MAFF
      Statistics on the Agricultural Labor Force: Population of Agricultural Workers and Number of Core Agricultural Workers, 2020.
      ). The World Health Organization defines old age as persons older than 65 years. According to this definition, the percentage of elderly people engaged in agriculture is extremely high compared with other industries (
      • MHLW
      Basic survey on wage structure in 2019: Summary of results by industry.
      ).
      Food import and export have increased considerably at an international level; however, the number of farm workers showed a 16% decrease globally between 2000 and 2019 (
      • FAO
      World Food and Agriculture: Statistical Yearbook 2020.
      ). Similarly, the population of people mainly engaged in farming in Japan has decreased from approximately 3.4 million in 1995 to approximately 2.2 million in 2018 (data from
      • MAFF
      Census of Agriculture and Forestry in Japan: Annual Statistics, 1995 to 2018.
      ). The northernmost prefecture, Hokkaido, is the main food production region in Japan, with a 206% food self-sufficiency rate on a production value basis in 2017 (
      • MAFF
      Prefectural Food Self-Sufficiency Rate, 2018.
      ). Dairy farming is a common activity in Hokkaido, accounting for approximately 54.4% of the total milk production in Japan (
      • MAFF
      Milk and Dairy Products Statistics, 2018.
      ). However, the number of dairy farms in Hokkaido has decreased in recent years, with only 20 new dairy farms replacing every 200 farms that close each year (
      • Hokkaido Prefecture
      Status of dairy farming business breakaway in Hokkaido.
      ).
      According to the
      • Japan Policy Council
      For a Growing 21st Century: Stop the Birthrate Decline—Rural Revitalization Strategy.
      , approximately 78% of the municipalities in Hokkaido would experience depopulation from 2010 to 2040. The depopulation of rural areas has negative effects on education, medical services, health and welfare systems, and other social services. The physical and mental health levels in rural areas are lower than those in urban areas in Australia and the United States (
      • Dixon J.
      • Welch N.
      Researching the rural-metropolitan health differential using the ‘Social Determinants of Health.’.
      ;
      • Hartley D.
      Rural health disparities, population health, and rural culture.
      ;
      • Smith K.B.
      • Humphreys J.S.
      • Wilson M.G.
      Addressing the health disadvantage of rural populations: How does epidemiological evidence inform rural health policies and research?.
      ), and similar observations have been made in Japan (
      • Nishi N.
      • Sugiyama H.
      • Kasagi F.
      • Kodama K.
      • Hayakawa T.
      • Ueda K.
      • Okayama A.
      • Ueshima H.
      Urban-rural difference in stroke mortality from a 19-year cohort study of the Japanese general population: NIPPON DATA80.
      ;
      • Kondo K.
      Social determinants of health and rural medicine.
      ).
      Particularly, the mental health level of dairy farmers is lower than that of workers in other industries (
      • Wallis A.
      • Dollard M.
      Local and global factors in work stress: The Australian dairy farming exemplar.
      ). Contributing factors to the poor mental health of dairy farmers include financial issues, irregular weather conditions owing to climate change, anxiety about the future of the farm, market price of farm products and livestock, tax, medical treatment fees, and lack of family time (
      • Momose Y.
      • Suenaga T.
      • Une H.
      Job satisfaction and mental distress among Japanese farmers.
      ;
      • Kearney G.D.
      • Rafferty A.P.
      • Hendricks L.R.
      • Allen D.L.
      • Tutor-Marcom R.
      A cross-sectional study of stressors among farmers in Eastern North Carolina.
      ;
      • Daghagh Yazd S.
      • Wheeler S.A.
      • Zuo A.
      Key risk factors affecting farmers' mental health: A systematic review.
      ).
      • Kanamori M.
      • Kondo N.
      Suicide and types of agriculture: A time-series analysis in Japan.
      reported that suicide rates were higher in areas with active livestock production in Japan; in addition, a stronger correlation was found between livestock production and suicide among men than among women; however, the reasons for the trends are yet to be clarified.
      To facilitate sustainable dairy farming, it is critical to assess and support the mental health of dairy farm workers. According to
      • Hansen B.G.
      • Østerås O.
      Farmer welfare and animal welfare—Exploring the relationship between farmer's occupational well-being and stress, farm expansion and animal welfare.
      , correlations exist among farmer welfare, farmer stress, and animal welfare.
      • King M.T.M.
      • Matson R.D.
      • DeVries T.J.
      Connecting farmer mental health with cow health and welfare on dairy farms using robotic milking systems.
      investigated relationships between the mental health of dairy farmers and milk production and udder health, in addition to indicators of farm animal welfare. Anxiety and depression were greater among farmers on farms that produced milk with lesser milk protein percentages. Research on the relationship between farm animal welfare and farmer welfare has been initiated, and some findings are beginning to emerge in the farm animal welfare research field. Dairy farms with different types of facilities are evaluated using indexes of physical strain and mental stress as part of occupational health and safety assessments in research on public health (
      • Karttunen J.P.
      • Rautiainen R.H.
      • Lunner-Kolstrup C.
      Occupational health and safety of Finnish dairy farmers using automatic milking systems.
      ).
      However, the limited investigations on mental health issues in dairy workers reduce the opportunities for recommendation of practical approaches to address the challenge. Therefore, the aim of this study was to examine the relationship between management factors and mental health among 81 dairy farm managers (80 male, 1 female) on selected dairy farms in Hokkaido, using quantitative surveys. Management factors include the number of livestock bred, the quantity of feed provided, farming materials invested, and milk production and revenue. An understanding of the major management factors that negatively affect the mental health of dairy farmers would facilitate formulation of policies and practical approaches to supporting dairy farm workers, and facilitate sustainable farming activities.

      MATERIALS AND METHODS

      Ethical Considerations

      This study was approved by the Ethics Committee of the Research Faculty of Agriculture, Hokkaido University, Sapporo, Japan, on October 4, 2018 (additional approval on November 8, 2018), to protect the rights of the participants in the study. The results of the study are accessible to farmers who provided written consent during the study.

      Selection of Study Site and Farms

      Despite the increased promotion of intensification, mechanization, and scaling of dairy management to improve productivity in Hokkaido, approximately 80% of the farms breeding dairy cattle older than 2 years old are family-run businesses with fewer than 80 cattle, and this is the mainstream management style (data from
      • MAFF
      Statistical Survey on Farm Management and Economy, Livestock Production Costs, Hokkaido, 2004–2015.
      ,
      • MAFF
      Statistical Survey on Farm Management and Economy, Statistics of Farming Types, 2004–2015.
      ). On average, there are 2.4 family farm workers and 73.7 cattle per farm in Hokkaido (
      • Kato H.
      • Kaichi K.
      • Morita S.
      Changes in the health condition of livestock with reductions of dairy farm workers.
      ).
      The study was conducted in town A and town B in Hokkaido. The key industry of town A is agriculture, where approximately 65% of farms engage in stockbreeding (dairy farming) and 35% in field farming, and the town is known as a field farm and dairy farm mixture area. The key industry of town B is also agriculture, but most farms engage in stockbreeding (dairy farming), and the town is known as a grassland dairy farming area. The research team collaborated with the Japan Agricultural Cooperative (JA; Chiyoda ward, Tokyo, Japan), which is an agricultural organization, in data-collection activities, based on the principle of mutual cooperation.
      A total of 100 farms in town A and 135 farms in town B belong to JA. We selected 42 farms in town A and 39 farms in town B for survey and data collection. Details of the selected farms are shown in Table 1.
      Table 1The number of dairy farms, cows, milk production value ratio, and dairy management details
      ItemUnitJapanHokkaidoTown ATown B
      Dairy farmsFarm15,700.0
      MAFF Hokkaido (2020).
      6,140.0
      MAFF Hokkaido (2020).
      100.0
      MAFF (2020b).
      189.0
      MAFF (2020b).
      CowsHerd/farm84.6
      MAFF Hokkaido (2020).
      128.8
      MAFF Hokkaido (2020).
      187.9
      MAFF (2020b).
      118.9
      MAFF (2020b).
      Ratio of milk production value in agricultural production value%8.3
      MAFF Hokkaido (2020).
      30.4
      MAFF Hokkaido (2020).
      48.3
      MAFF Hokkaido (2020).
      74.4
      MAFF Hokkaido (2020).
      Number of study farmsFarm4239
      Milking type
       PipelineFarm2029
       Milking parlor or automatic milking systemFarm2210
      Barn type
       Stanchion barnFarm1824
       Loose barnFarm2415
      1 MAFF
      • Hokkaido M.A.F.F.
      Annual Report of Statistics on Agriculture, Forestry and Fisheries of Hokkaido 2018-2019.
      .
      2
      • MAFF
      Agriculture, Forestry, and Fisheries in Graphs and Statistics, 2020.12.17update ver.
      .

      Research Items: Defining the Management Factors of Dairy Production Systems

      Management factors are the essential elements influencing a dairy production system, such as the number of livestock bred, the quantity of feed provided, farming materials invested, milk production and revenue, and so on. In this study, management analysis was conducted using a mainly financial table, which was divided into 3 categories: input, output, and facility, with a total of 18 key management-related variables in the categories V01 to V18. These items were used to examine the management factors of the dairy farmers. Table 2 shows an itemization of managerial factors in this study.
      Table 2Descriptive statistics of the study farms by 2 different milking systems
      Variable numberVariable description
      AU = animal units; MFTA = medical consultation fee per unit time per AU; FPCM = fat- and protein-corrected milk; LS = milk quality variable; CES-D = Center for Epidemiologic Studies Depression Scale.
      UnitAll farms (n = 81)Pipeline milking (n = 49)Milking parlor or AMS
      AMS = automatic milking system.
      (n = 32)
      P-value
      NS = not significant.
      MeanSDCVMeanSDMeanSD
      V01Farmlandha64.850.00.7753.226.582.669.40.028
      P < 0.05,
      V02Cows by AUAU145.7119.30.8292.868.8226.5135.00.000
      P < 0.001.
      V03
      Principal component analysis variables.
      Stocking densityAU/ha2.72.60.962.12.13.73.10.008
      P < 0.01,
      V04
      Principal component analysis variables.
      Concentrated feedkg/milking cow per year2,851.61,301.50.462,636.51,156.43,180.81,454.60.065NS
      V05
      Principal component analysis variables.
      Nonfamily wages1,000 yen/AU46.284.01.8222.567.580.994.20.004
      P < 0.01,
      V06
      Principal component analysis variables.
      Fertilizer and pesticide expenses1,000 yen/ha88.875.10.8574.273.2111.373.50.029
      P < 0.05,
      V07Electricity, water, gas, and fuel expenses1,000 yen/ha148.0160.21.08111.6139.7203.7175.30.011
      P < 0.05,
      V08
      Principal component analysis variables.
      Livestock care cost1,000 yen/AU66.055.20.8454.347.983.861.40.018
      P < 0.05,
      V09Number of livestock medical visitsTimes/AU1.91.10.611.81.22.01.10.552NS
      V10Livestock medical fees1,000 yen/AU21.713.90.6425.116.016.77.50.002
      P < 0.01,
      V11
      Principal component analysis variables.
      MFTAYen/times per AU172.0160.90.94224.3156.791.9133.70.000
      P < 0.001.
      V12
      Principal component analysis variables.
      FPCMkg/milking cow per year8,922.31,607.50.188,443.31,576.69,655.71,379.10.001
      P < 0.001.
      V13Agricultural income1,000 yen/AU1,323.7681.40.511,216.5746.61,487.7538.00.080NS
      V14Agricultural expenses1,000 yen/AU1,007.5569.10.56901.9611.71,169.3460.00.038
      P < 0.05,
      V15Net agricultural income1,000 yen/AU316.1169.00.53314.6175.5318.4161.10.921NS
      V16
      Principal component analysis variables.
      Net agricultural income ratio%25.18.40.3327.28.322.17.60.006
      P < 0.01,
      V17
      Principal component analysis variables.
      LS2.40.50.192.50.52.40.40.209NS
      V18
      Binomial logistics regression analysis variable.
      CES-D total score12.68.90.7012.610.212.56.50.960NS
      1 AU = animal units; MFTA = medical consultation fee per unit time per AU; FPCM = fat- and protein-corrected milk; LS = milk quality variable; CES-D = Center for Epidemiologic Studies Depression Scale.
      2 AMS = automatic milking system.
      3 NS = not significant.
      4 Principal component analysis variables.
      5 Binomial logistics regression analysis variable.
      * P < 0.05,
      ** P < 0.01,
      *** P < 0.001.

      Management Factors in Dairy Production Systems Index 1: Input Indicators

      Production-Based Indicators

      The production-based indicators were stocking density [animal units (AU)/ha, V03] and the amount of concentrate feed provided (kg/milking cow per year, V04). The stocking density was obtained by dividing the management area (ha, V01) by the AU (V02). The management area was the total grassland, pastureland, grass crop fields, leased land, and other crop fields, excluding residential areas, mountains, forests, and leasehold land provided to a tenant. The AU was determined based on the AU coefficient for livestock bred on farms (
      • Kato H.
      • Kaichi K.
      • Morita S.
      Changes in the health condition of livestock with reductions of dairy farm workers.
      ). The amount of concentrated feed provided (kg/milking cow per year) was the amount of concentrated feed provided to a milking cow per year. The variables in this study were converted into units per milking cow, AU, and management area for clarity.

      Nonfamily Wage Indicators

      In dairy production systems, the level of dependency on external labor is a variable that has significant influence on both economic efficiency and the workload of family farm workers. The variable includes nonfamily wages (yen/AU, V05). For farming households for which nonfamily wages could not be separated from the total labor cost (11 households), the average value of the ratio of nonfamily wages to the total labor cost for each milking type was calculated and corrected to obtain the value of nonfamily wages. The proportions of nonfamily wages within the total labor cost were 17, 71, and 58%, representing pipeline, parlor, and automatic milking system (AMS), respectively. The value of 100 Japanese yen (US$0.92) in March 12, 2020, was used for wage calculation.

      Environmental Load Indicators

      The demand for environmentally conscious production systems is considerable. Taking this into account, we included fertilizer and pesticide expenses per unit area (yen/ha, V06) and electricity, water, gas, and fuel expenses per unit area (yen/ha, V07) as variables in this study. These expenses were considered to represent environmental loads, which are sufficient to evaluate relative impact in our statistical analysis.

      Livestock Disease Indicators

      In dairy production systems, livestock disease significantly influences both economic efficiency and milk production. Parameters such as livestock care cost (yen/AU, V08), number of medical visits per livestock unit (times/AU, V09), and livestock medical fees (yen/AU, V10) were set as livestock disease indicators in this study. Livestock care costs included sanitation expenses, artificial insemination fees, deposit fees, litter fees, and hoof cutting fees. Livestock medical fees, including both medical and technical fees, are expressed as medical consultation fee per unit time per AU (MFTA, yen/times per AU, V11). The criterion for being visited or not visited by a veterinarian varies among farm managers; therefore, it may not be an ideal factor for expressing the current situation on the farm. However, livestock medical fees and number of livestock medical visits were considered adequate to indicate the comprehensive livestock health, and in turn, the major farm animal welfare indicators (
      • Kato H.
      • Kaichi K.
      • Morita S.
      Changes in the health condition of livestock with reductions of dairy farm workers.
      ).

      Management Factors in Dairy Production Systems Index 2: Output Indicators

      Production Indicators: Fat- and Protein-Corrected Milk

      The major management goal of dairy production system is the sustainable production of fresh milk. Therefore, milk output was used as the productivity index in this study. Milk production was calculated based on the value of fat- and protein-corrected milk (FPCM, V12;
      • IDF
      A common carbon footprint approach for dairy: The IDF guide to standard lifecycle assessment methodology for the dairy sector. Bull. Int. Dairy Fed.
      ). The FPCM is a system that corrects the variability of fat and protein content of produced milk to standardized ratios of 4.0% and 3.3%, respectively, and is used to compare milk with varying fat and protein contents. The FPCM is calculated using the following equation:
      FPCM (kg/yr) = Production × (0.1226 × Fat% + 0.0776 × True protein% + 0.2534),


      where production is in kilograms per year.

      Economic Indicators

      We also examined the economic sustainability of the dairy production systems. The indicator used for this was collected from the farm accounting table, known as KUMIKAN, in Japan. Japan's KUMIKAN is a network of microeconomic data based on bookkeeping principles similar to the Farm Accountancy Data Network used in the EU.
      Data captured in KUMIKAN are limited to the content of transactions between farms and JA, and does not keep track of transactions that do not go through JA. The depreciation costs of equipment and facilities are also not included; however, they are useful in analyzing the economy. Agricultural income (yen/AU, V13), agricultural expenses (yen/AU, V14), net agricultural income (yen/AU, V15), and net agricultural income ratio (%, V16) were calculated using KUMIKAN data. Net agricultural income was obtained by subtracting agricultural expenses from agricultural income. The net agricultural income ratio was the ratio of net agricultural income to agricultural income, which indicates economic efficiency. In this study, the net agricultural income ratio (%, V16) was adopted as the productivity index variable:
      Net agricultural income (1,000 yen) = Agricultural income – Agricultural expenses;


      Netagriculturalincomeratio(%)=NetagriculturalincomeAgriculturalincome×100.


      Milk Quality Indicators

      The quality of milk produced has a considerable effect on the profitability of dairy production systems. Japanese dairy manufacturers set the market price of milk based on the somatic cell count of fresh milk, nonfat milk solids ratio, and butterfat (
      • Snow Brand Seed Co
      How milk prices are paid by dairy manufacturers.
      ). In this study, the average somatic cell count (V17) was used as the milk quality variable (LS). Somatic cell count is a general term for the concentration of white blood cells and shed epithelial cells in milk, and is characteristically high in mastitis-infected cattle. Herd and population somatic cell counts are related to the inflammatory process in individual cows. They reflect the udder health status of the herd and the quality of the raw milk in the herd and the population (
      • Schukken Y.H.
      • Wilson D.J.
      • Welcome F.
      • Garrison-Tikofsky L.
      • Gonzalez R.N.
      Monitoring udder health and milk quality using somatic cell counts.
      ). The LS is obtained by taking the logarithm of the somatic cell count per 1 mL of fresh milk produced by milking cows and is graded from 0, as the optimal milk quality grade, to 9 (
      • Aihara M.
      Today and tomorrow, the dairy herd milk records promise to help your dairy farm management.
      ). In Japan, the management goal is to maintain the number of herds with grade 5 or higher (SCC of 283,000–565,000 cells/mL) at less than 8% in all herds.

      Management Factors in Dairy Production Systems Index 3: Facility Indicators

      The milking type and feeding method of each farm were used as the facility variables. Each variable was converted into a dummy variable. The milking types were pipeline and milking parlor or AMS and were assigned scores of 0 and 1, respectively. Tiestall and freestall or free-barn feeding methods were assigned scores of 0 and 1, respectively. The feeding facilities of 83.7% of pipeline farms were tiestalls, and those of 96.7% of milking parlor or AMS farms were freestall or free barns. In this study, milking type was used as the facility index of the farms.
      Data from 2018 were used to examine the management factors of dairy farms. Data provided by JA were used as key items associated with business overview and economic efficiency. Information on farm livestock diseases was based on livestock medical records provided by the National Agricultural Insurance Association. Variables associated with milk quality were based on data from the dairy herd improvement program by the Hokkaido Dairy Milk Recording and Testing Association.

      Farmers' Mental Health (Index of Farmer Welfare): Depressive Symptoms

      Mental health is increasingly a source of concern as an occupational and widespread issue in Japan. According to the Ministry of Health, Labor, and Welfare (
      • MHLW
      Overview of the estimation of economic benefits of suicide and depression prevention measures (Social losses due to suicide and depression).
      ), the economic losses attributed to mental health issues amount to approximately 2.7 trillion yen in 2009. Particularly, depression, as one of the major diseases, tends to be lifelong, and it is used extensively in medical services as a measure of workers' mental health. Depression affects mental and physical health and, as the most common mental illness, is an important public health problem (
      • Liu Q.
      • He H.
      • Yang J.
      • Feng X.
      • Zhao F.
      • Lyu J.
      Changes in the global burden of depression from 1990 to 2017: Findings from the Global Burden of Disease study.
      ). In the present study, the mental health status of farm workers was evaluated based on depression systems using the Center for Epidemiologic Studies Depression Scale (CES-D, V18), a 20-item assessment that is commonly used to screen for depression. The CES-D has been translated into more than 30 languages, and its reliability and validity confirmed in the Japanese version (
      • Shima S.
      • Shikano T.
      • Kitamura T.
      • Asai M.
      New self-rating scale for depression.
      ). The CES-D has also been applied in mental health research among farmers (
      • Yazd S.D.
      • Wheeler S.A.
      • Zuo A.
      Key risk factors affecting farmers' mental health: A systematic review.
      ).
      In the CES-D, each question item asks how often respondents have experienced the symptoms within the last week, and responses are based on a scale ranging from 0 to 3: for instance, 0 = rarely or none of the time and 3 = most or all the time (5–7 d). All responses are added to obtain a total score; the potential range of the total score is 0 to 60. A cut-off point of 15 or 16 is commonly used to identify those with depressive or those with non-depressive symptoms on the CES-D scale. In the present study, participants with a total score of 0 to 15 were categorized as non-depressive, and those with total scores of 16 to 60 were considered as having depressive symptoms.
      For the analyses, each cut-off value was converted into a dummy variable; depression and non-depression were assigned scores of 1 and 0, respectively. This was a cross-sectional study. The survey on the mental health of 81 farm managers (80 male, 1 female) was conducted via questionnaires from December 2018 to February 2019, and the CES-D score of the farm manager was used to determine the level of depression.

      Research Framework and Statistical Analysis

      A basic statistical analysis of the variables was conducted, and a t-test was used to compare the mean values of 2 populations with 2 different milking methods: pipeline as conventional milking type, and parlor and AMS as semi-automatic or automatic milking type (Table 2). Parlor and AMS milking methods have less quantitative workload than pipeline milking. Mean values were considered significant at P < 0.05.
      Subsequently, Pearson correlation analysis was performed to investigate the relationships between the variables. Correlation was considered significant for 2-tailed at the levels of P < 0.05 or P < 0.01. Variables with a Pearson correlation coefficient and coefficient of variation of less than 0.8 and more than 0.1, respectively, were applied in a principal component analysis (PCA;
      • Pearson K.
      On lines and planes of closest fit to systems of points in space.
      ). Principal component analysis is a technique that creates new synthetic variables from multiple data variables, to facilitate interpretation of data. In this study, the principal factor method of PCA was used to aggregate and interpret management factors.
      The PCA calculated the principal components, which are new variables constructed as linear combinations of the initial variables. The standardized factor loading should be greater than 0.5 for interpretability criteria (
      • Hair Jr., J.F.
      • Black W.C.
      • Babin B.J.
      • Anderson R.E.
      • Tatham R.L.
      Multivariate Data Analysis.
      ). Therefore, in this study, the variables with factor loadings greater than 0.5 or less than −0.5 in each principal component were considered essential interpretability criteria. Simultaneously, each farm's principal component score was calculated for component eigenvalues greater than 1.0.
      Associations between the variables and the mental health of farm managers were assessed using binary logistic regression analysis (
      • Cox D.R.
      The regression analysis of binary sequences.
      ), with the CES-D cut-off value as the dependent variable and the principal component score and facility index variable as the explanatory variables. All data analyses were performed using SPSS Statistics version 27 (IBM Corp.).

      RESULTS AND DISCUSSION

      Descriptive Statistics

      Preliminary statistics for the variables used in the study are presented in Table 2. We found no significant difference in the number of livestock medical visits; however, the MFTA was significantly higher in the pipeline system than in parlor or AMS systems. In the present study, data on the number of livestock medical visits and medical fees were based solely on the consultation record of veterinarians, without consideration of on-farm treatments. Although livestock care cost was included in this study as a livestock disease indicator, it will also be necessary to evaluate farm animal welfare based on physiological indices and housing facility conditions in the future. Regarding the economic efficiency of the 2 systems, we observed no significant differences between the agricultural income (P = 0.080) and net agricultural income (P = 0.921) generated in both systems. However, parlor and AMS milking had significantly higher (P = 0.038) agricultural expenses compared with the pipeline system, whereas the pipeline system had a significantly higher net agricultural income ratio. This is an indication that the pipeline system is a more economically efficient system despite its low total agricultural income. No significant difference was detected between the total CES-D (P = 0.960) scores of farm managers in the 2 systems.

      Correlation Matrix

      The results of the correlation analysis of the variables are shown in Table 3. The AU (r = 0.625; P < 0.01), quantity of concentrate feed provided (r = 0.307; P < 0.01), nonfamily wages (r = 0.541; P < 0.01), fertilizer and pesticide expenses (r = 0.514; P < 0.01), electricity, water, gas, and fuel expenses (r = 0.843; P < 0.01), livestock care cost (r = 0.658; P < 0.01), FPCM (r = 0.635; P < 0.01), and number of livestock medical visits (r = 0.333; P < 0.01) were significantly positively correlated with stocking density. The higher the stocking density, the higher the quantity of inputs for production. A strong positive correlation was detected between agricultural income (r = 0.519; P < 0.01), agricultural expenses (r = 0.601; P < 0.01), and stocking density; and a weak positive correlation (r = 0.262; P < 0.05) was found between net agricultural income and stocking density. The net agricultural income ratio was significantly negatively correlated (r = −0.449; P < 0.01) with stocking density, indicating that highly intensive dairy management systems are not always economically efficient. Livestock medical fees (1,000 yen/AU; r = −0.378; P < 0.01), MFTA (yen/times per AU; r = −0.760; P < 0.01), and LS (r = −0.290; P < 0.01) were significantly negatively correlated with stocking density.
      Table 3Pearson correlation matrix on all dairy management factor variables (n = 81)
      Variable number and definition
      AU = animal units; MFTA = medical consultation fee per unit time per AU; FPCM = fat- and protein-corrected milk; LS = milk quality variable; CES-D = Center for Epidemiologic Studies Depression Scale. Correlation is significant for 2-tailed at *P < 0.05 or **P < 0.01.
      UnitV01V02V03V04V05V06V07V08V09V10V11V12V13V14V15V16V17V18
      V01: Farmlandha1
      V02: Cows by AUAU0.507**1
      V03: Stocking densityAU/ha−0.243*0.625**1
      V04: Concentrated feedkg/milking cow per year−0.1740.1700.307**1
      V05: Nonfamily wages1,000 yen/AU0.0150.513**0.541**0.1781
      V06: Fertilizer and pesticide expenses1,000 yen/ha−0.236*0.227*0.514**0.0640.531**1
      V07: Electricity, water, gas and fuel expenses1,000 yen/ha−0.332**0.456**0.843**0.325**0.621**0.703**1
      V08: Livestock care cost1,000 yen/AU−0.261*0.374**0.658**0.2170.487**0.622**0.705**1
      V09: Number of livestock medical visitsTimes/AU−0.1750.1320.333**−0.0040.309**0.527**0.447**0.471**1
      V10: Livestock medical fees1,000 yen/AU0.191−0.186−0.378**−0.001−0.252*−0.413**−0.425**−0.273*0.271*1
      V11: MFTAYen/times per AU−0.080−0.743**−0.760**−0.094−0.623**−0.648**−0.757**−0.627**−0.468**0.507**1
      V12: FPCMkg/milking cow per year−0.1250.444**0.635**0.480**0.504**0.427**0.662**0.571**0.332**−0.151−0.525**1
      V13: Agricultural income1,000 yen/AU−0.446**0.0810.519**0.289**0.555**0.780**0.784**0.659**0.475**−0.388**−0.537**0.641**1
      V14: Agricultural expenses1,000 yen/AU−0.399**0.1920.601**0.283*0.598**0.784**0.824**0.711**0.529**−0.327**−0.605**0.678**0.956**1
      V15: Net agricultural income1,000 yen/AU−0.486**−0.1520.262*0.334**0.282*0.573**0.499**0.432**0.167−0.461**−0.262*0.401**0.757**0.593**1
      V16: Net agricultural income ratio%0.032−0.395**−0.449**0.011−0.448**−0.422**−0.505**−0.420**−0.496**−0.0080.513**−0.330**−0.363**−0.578**0.242*1
      V17: LS0.060−0.234*−0.290**−0.209−0.215−0.107−0.191−0.1910.0780.277*0.210−0.263*−0.124−0.1030.1850.0651
      V18: CES-D total score0.0580.030−0.0490.027−0.0450.013−0.0360.047−0.0300.1100.017−0.0140.0130.0110.0250.0090.0501
      1 AU = animal units; MFTA = medical consultation fee per unit time per AU; FPCM = fat- and protein-corrected milk; LS = milk quality variable; CES-D = Center for Epidemiologic Studies Depression Scale.Correlation is significant for 2-tailed at *P < 0.05 or **P < 0.01.
      The replacement time for cows is usually short under intensive management of high-producing dairy cows (
      • Ohgi T.
      • Shiga E.
      Consider the number of breeding years of a dairy cow—Its facts and determinants.
      ). This is due to increases in occurrence of illness (
      • Ohgi T.
      • Shiga E.
      Consider the number of breeding years of a dairy cow—Its facts and determinants.
      ) and somatic cell counts of fresh milk (
      • Sasano M.
      Quality Control of Raw Milk that Promises Safety and Security to Consumers.
      ) as the parity increases.
      • Ohgi T.
      • Kaneko T.
      • Doukoshi A.
      • Kusakari N.
      • Hatta T.
      Effects of milk yield and feeding management on disposal and disease in dairy herds in Hokkaido.
      reported that farms with high-producing dairy cows were more willing to cull cows with mastitis and reproductive disorders than farms with low-producing dairy cows. Dairy farming is privately controlled, and, therefore, farmers maintain profitability and milk quality by culling sick animals to cut down on livestock medical expenses. This may not be considerate for the welfare of the farm animal. Such a practice in the livestock industry that does not take into account farm animal welfare could affect farm sustainability in the long term. We found no significant correlation between the total CED-D score (V18) and the variables. In a previous study, no association was found between milk yield, somatic linear score, and farmer depression (
      • King M.T.M.
      • Matson R.D.
      • DeVries T.J.
      Connecting farmer mental health with cow health and welfare on dairy farms using robotic milking systems.
      ). The scales used to assess mental health in this study were different from those used in the earlier study; however, similar results were obtained for the single-correlation relationship, with no correlation between milk yield, LS, and CES-D depression symptoms.

      Integration of Management Factors in Dairy Production Systems Index: PCA

      Variables with Pearson correlation coefficients and coefficients of variation less than 0.8 and more than 0.1, respectively, were used for PCA. A total of 10 representative variables were selected initially. Electricity, water, gas, and fuel expenses per farmland had a strong correlation (r = 0.842; P < 0.01) with stocking density, and, therefore, they were combined to form one variable. Nine variables were used for PCA (Table 4).
      Table 4Principal component (PC) factor loadings for 9 dairy management factor variables
      AU = animal unit; FPCM = fat- and protein-corrected milk; MFTA = medical consultation fee per unit time per AU; LS = milk quality variable.
      ItemUnitPC1PC2
      Variable number and definition
       V08: Livestock care costYen/AU0.788
      Loadings with absolute value greater than ±0.5.
      −0.002
       V12: FPCMkg/milking cow per year0.715
      Loadings with absolute value greater than ±0.5.
      0.321
       V03: Stocking densityAU/ha0.715
      Loadings with absolute value greater than ±0.5.
      0.341
       V11: MFTAYen/times per AU−0.681
      Loadings with absolute value greater than ±0.5.
      0.313
       V05: Nonfamily wagesYen/AU0.633
      Loadings with absolute value greater than ±0.5.
      −0.221
       V06: Fertilizer and pesticide expensesYen/ha0.610
      Loadings with absolute value greater than ±0.5.
      −0.480
       V16: Net agricultural income ratio%−0.533
      Loadings with absolute value greater than ±0.5.
      0.528
      Loadings with absolute value greater than ±0.5.
       V04: Concentrated feedkg/milking cow per year0.4130.633
      Loadings with absolute value greater than ±0.5.
       V17: LS−0.337−0.579
      Loadings with absolute value greater than ±0.5.
      Eigenvalue3.4491.611
      Contribution rate%38.317.9
      Cumulative contribution rate%38.356.2
      1 AU = animal unit; FPCM = fat- and protein-corrected milk; MFTA = medical consultation fee per unit time per AU; LS = milk quality variable.
      * Loadings with absolute value greater than ±0.5.
      The results of PCA showed that the cumulative contribution rate of the variables in principal components 1 and 2 (PC1 and PC2) was 56.2%. Livestock care cost, FPCM, stocking density, MFTA, nonfamily wages, fertilizer and pesticide expenses, and net agricultural income ratio were clustered in PC1, with a contribution rate of 38.3%. Variables with factor loadings >0.7, including livestock care cost, FPCM, and stocking density, were extracted as positive loadings. High factor loadings indicate a high correlation between the assessment items and each factor. Management systems with high FPCM and stocking density are usually highly intensive. Therefore, variables clustered in PC1 were classified as “intensiveness factors.”
      Net agricultural income ratio, quantity of concentrate feed, and LS were clustered at PC2, with a contribution rate of 17.9%. Management factors in PC2 were basic; one factor is related to feeding, one is related to finances in general, and the last is related to milk quality. Therefore, PC2 was classified as “basic dairy management factors.” Considering the factor loadings in PC2, it was found that a high net agricultural income ratio, which indicates a positive loading rate, leads to stable economic farm management. High concentrate feed quantity, indicating a positive loading rate, implies adequate feed supply to cows. Low LS had a negative loading rate and indicated high milk quality. Thus, PC2 could also be expressed as “well-managed farm systems.” Table 2 shows no statistical mean difference in the quantity of concentrate feed and LS, between the different milking systems, or in other words, between different farm sizes. These 2 variables are considered as management factors dependent more on the farm managers' decisions than the milking system or farm scale. Therefore, PC2 consists of dominant management factors influenced by farm managers' decisions.
      The factor loadings for net agricultural income ratio in PC1 were negative, which indicates that the farms with low net agricultural income ratio are highly intensive farms. The net agricultural income ratio of PC2 had a positive factor loading. Farms with higher net agricultural income ratio are associated with feeding and milk quality, which indicates the latent meaning of the PC2 as “well-managed farm systems.” The agricultural income ratio influenced PC1 and PC2, indicating that economic efficiency was a key factor in assessing management conditions.

      Relationship Between Management Factors and Depression Symptoms of Farm Managers: Binomial Logistic Regression Analysis

      A binary logistic regression analysis was performed using the cut-off value of CES-D as the dependent variable, and PC1, PC2, and facility index variable as independent variables (Table 5). A P-value of 0.445 in the Hosmer–Lemeshow goodness of fit test indicated that the model was well-fitted. In addition, the discriminant contingency table correctly predicted 78.5% of the subjects. The highest odds ratio in PC2 (2.237), compared with those in PC1 and milking type (0.870 and 1.909, respectively) indicated it had the greatest influence on depressive symptoms. However, the depressive symptoms of farm managers was not significantly associated with intensiveness factors (PC1; P = 0.684) and milking type (P = 0.307). In contrast, PC2 score was significantly related (P = 0.018) to depressive symptoms, which indicated that farm managers with depressive symptoms had superior dairy management factors, which were evaluated based on net agricultural income ratio, concentrated feed provided, and linear score of somatic cell count. An increase in PC2 score was associated with depression conditions; and, therefore, we hypothesized that better dairy management had positive effects on mental health. However, the results were inconsistent with our expectations.
      Table 5Binomial logistic regression analysis of 3 factors relating to depressive symptoms
      Variable
      PC = principal component.
      B
      B = coefficient of the model equation.
      SE of BOdds ratio95% CIP-value
      NS = not significant.
      LowerUpper
      PC1 score (intensiveness factors)0.1390.3040.8700.4791.5800.648NS
      PC2 score (basic dairy management factors/well-managed farm systems)0.8050.3412.2371.1484.3620.018
      P < 0.05.
      Milking system type
      Reference category is 0 as pipeline milking system.
      0.6470.6341.9090.5526.6080.307NS
      1 PC = principal component.
      2 B = coefficient of the model equation.
      3 NS = not significant.
      4 Reference category is 0 as pipeline milking system.
      * P < 0.05.
      Psychological and physical subjective symptoms, including anxiety and depression symptoms, are considered as psychological reactions that are part of the acute stress response and are reported as theoretical models of work stress (
      • Hurrell Jr., J.J.
      • McLaney M.A.
      Exposure to job stress—A new psychometric instrument.
      ;
      • Araki S.
      • Kawakami N.
      Health care of work stress: A review.
      ). In this section, we discuss the stressors in dairy management that bring about depression symptoms. The targets of this study are the farm managers. Mental health problems are observed mainly in managers involved in their company, due to financial problems and the demands of running the business continuously (
      • Ishino S.
      • Matsuoka H.
      • Yamada J.
      • Ogasawara E.
      • Takeuchi K.
      • Bumsuk L.
      • Shiihara Y.
      Manager's attitude toward mental health care in small and medium-sized enterprises (I)—A survey by semi-structured interviews.
      ;
      • Sano S.
      • Tanaka K.
      A study of job stressors and stress coping strategies among managers in small and medium-sized enterprises.
      ). Managers could be under more financial stress than employees, because they have to ensure sustained financial management. We discuss how each dairy management factor, such as agricultural income ratio, concentrated feed provided, and LS, is a stress factor and how they are related in dairy management. Dairy farmers are under pressure to maintain high milk quality, to avoid fines in the event of supplying fresh milk with high somatic cell count (
      • Kurihara S.
      • Shimoura S.
      Quality control efforts among Japanese dairy farmers: Findings from national field surveys.
      ). They have to pay great attention to feeding approaches, including nutrient balance and feeding methods, because, for example, roughage ratio in animal feed influences milk components (
      • Akuzawa R.
      • Ishibashi A.
      Animal Feed Science. Vol. 61. Animal Husbandry. No. 43.
      ;
      • Moorby J.M.
      • Lee M.R.
      • Davies D.R.
      • Kim E.J.
      • Nute G.R.
      • Ellis N.M.
      • Scollan N.D.
      Assessment of dietary ratios of red clover and grass silages on milk production and milk quality in dairy cows.
      ). Individual daily hygiene management of dairy cows requires that farmers expend considerable mental effort to keep the udders clean and healthy, and in facility management, which could greatly influence milk quality (
      • Japan Dairy Council
      Guideline for Good-Quality Milk Production.
      ).
      • Kawai K.
      • Kurosawa S.
      • Nagahata H.
      Relationship between milking management practices and milk somatic cell counts on local dairy farms.
      reported that milk hygiene, which affects somatic cell count, is determined by the performance of dairy cows during the dry period, timing of implementation of post-dipping, degree of overmilking, dip type, use of milking gloves, and method of injection of ointment during mastitis treatment. Overall, milking activity should be performed under hygienic conditions and with care. The factors are interrelated with each other, and they influence each other; it is difficult improve each of them in isolation. In our PCA, PC2 indicates the integrated influence of the 3 factors. Therefore, the important point in this study is “the simultaneous nature of the demands.” This is probably because the maintenance of proper financial status, feeding environment for cows, and milk quality simultaneously lead to high psychological stress levels for farm managers. Therefore, it can be inferred that the application of appropriate techniques to run well-managed farm systems requires considerable effort and sacrifices from farm workers and managers, which can cause and increase stress. In other words, the simultaneous nature of the demands cannot be measured simply in the form of good business conditions or low LS. Furthermore, the present research is based on an observational study, so that it cannot make conclusions regarding causation. In future, it will be necessary to examine the causality of the present results in a research framework that includes variables such as psychological stress and actual physical and mental exertion.
      However, we suggest that a practical approach for improving mental health on farm is “taking measures of sustaining the simultaneous nature of the demands in addition to stable economic farm management, adequate and rapid feed supply, and milk quality maintenance.” The results from this study could be helpful for supporting dairy farm management and future research.

      Limitations of the Study and Future Work

      This study was conducted in only 2 towns in Hokkaido and is not necessarily a true representation of the status in the entire country. Furthermore, questionnaires were administered only to farm managers, without considering other farm workers and gender differences. Over the years, there has been an increase in the participation of women in different sectors and industries of the economy, including agriculture. According to
      • MAFF
      Office for Promotion of Women's Activities, Promotion of Women's Activities in Agriculture.
      , approximately half of the population involved in agriculture are women. In addition, women account for 24.0% of all new farm workers, and the ratio of the female population among all core persons mainly engaged in farming is highest among those in their fifties, indicating that women play an important role in agricultural management. Moreover, in addition to wage-earning jobs, women in Japan often have family responsibilities, such as domestic chores and childcare (
      • Tsuru E.
      The Sociology of Farm Women: The Energy of Agriculture Starts with Women.
      ). Future studies should examine the mental health of female dairy farm workers.

      CONCLUSIONS

      This study examined the relationship between management factors and the mental health of dairy farm managers. The findings of this study showed that depression symptoms of dairy farmers were significantly negatively associated with the application of basic dairy management factors, including stable economic farm management, adequate and rapid feed supply, and milk quality maintenance. The limitations inherent in these factors could increase the depression levels of dairy farm managers and negatively affect their mental health, suggesting that sustaining the simultaneous nature of the demands requires overworking. Measures should be put in place to reduce physical and psychological stress among dairy farm workers for improved mental health. Improved mental health of the farmer has a positive effect on farm animal welfare. Unfortunately, in Japan, the term “farm animal welfare” is not well known, and farmers and consumers are unaware of it. Improving mental health of farmers could lead to positive changes in farmers' behavior toward livestock. This could improve the dairy production system in Japan and simultaneously address farm animal welfare, therefore enabling implementation of a more sustainable production system. Improving the mental health of agricultural employees should be emphasized for sustainable agriculture. There is a need to formulate policies and establish social systems that consider the physical and psychological health of dairy farm workers.

      ACKNOWLEDGMENTS

      We acknowledge the Japan Agricultural Cooperative, Hokkaido Dairy Milk Recording and Testing Association (Sapporo, Japan), and the National Agricultural Insurance Association (Tokyo, Japan) for their support. We also thank all participants for their willingness to participate in our study. We thank Editage (www.editage.com) for English-language editing. This research was funded by a Grant-in-Aid for Scientific Research (C) as part of the project on “Evaluation of the dairy farming systems for sustainability using multiple management and health science indicators” from the Japan Society for the Promotion of Science (Tokyo), grant number 18K05917. Conceptualization: H.K., O.H., M.S., and M.N.; data curation: H.K.; formal analysis and interpretation of data: H.K., O.H., M.S., and M.N.; funding acquisition: H.K.; investigation: H.K., O.H., M.S., and M.N.; methodology: H.K., O.H., and M.S.; project administration: H.K. and H.O.; initial draft of the paper: H.K.; supervision: H.O. and K.K. All authors reviewed and edited the manuscript and approved the final version submitted for publication. All authors have read and agreed to the published version of the manuscript. The authors have not stated any conflicts of interest.

      REFERENCES

        • Aihara M.
        Today and tomorrow, the dairy herd milk records promise to help your dairy farm management.
        Daly Japan Ltd, 2013 (ISBN-13: 974-4-9245506-65-7)
        • Akuzawa R.
        • Ishibashi A.
        Animal Feed Science. Vol. 61. Animal Husbandry. No. 43.
        https://agriknowledge.affrc.go.jp/RN/2010752019.pdf
        Date: 2007
        Date accessed: March 29, 2021
        • Araki S.
        • Kawakami N.
        Health care of work stress: A review.
        Jpn. J. Ind. Health. 1993; 35: 88-97
        • Cox D.R.
        The regression analysis of binary sequences.
        J. R. Stat. Soc. Series B Stat. Methodol. 1958; 20: 215-232
        • Daghagh Yazd S.
        • Wheeler S.A.
        • Zuo A.
        Key risk factors affecting farmers' mental health: A systematic review.
        Int. J. Environ. Res. Public Health. 2019; 16 (31810320)4849
        • Dixon J.
        • Welch N.
        Researching the rural-metropolitan health differential using the ‘Social Determinants of Health.’.
        Aust. J. Rural Health. 2000; 8 (11894255): 254-260
        • FAO
        World Food and Agriculture: Statistical Yearbook 2020.
        (Pages 4–19.)
        • Hair Jr., J.F.
        • Black W.C.
        • Babin B.J.
        • Anderson R.E.
        • Tatham R.L.
        Multivariate Data Analysis.
        7th ed. Prentice-Hall, 2009
        • Hansen B.G.
        • Østerås O.
        Farmer welfare and animal welfare—Exploring the relationship between farmer's occupational well-being and stress, farm expansion and animal welfare.
        Prev. Vet. Med. 2019; 170104741
        • Hartley D.
        Rural health disparities, population health, and rural culture.
        Am. J. Public Health. 2004; 94 (15451729): 1675-1678
        • Hokkaido Prefecture
        Status of dairy farming business breakaway in Hokkaido.
        • Hurrell Jr., J.J.
        • McLaney M.A.
        Exposure to job stress—A new psychometric instrument.
        Scand. J. Work Environ. Health. 1988; 14 (3393871): 27-28
        • IDF
        A common carbon footprint approach for dairy: The IDF guide to standard lifecycle assessment methodology for the dairy sector. Bull. Int. Dairy Fed.
        http://www.ukidf.org/documents/bulletin445.pdf
        Date: 2010
        Date accessed: December 20, 2020
        • Ishino S.
        • Matsuoka H.
        • Yamada J.
        • Ogasawara E.
        • Takeuchi K.
        • Bumsuk L.
        • Shiihara Y.
        Manager's attitude toward mental health care in small and medium-sized enterprises (I)—A survey by semi-structured interviews.
        Japn. J. Occup. Med. Traumatol. 2009; 57: 251-257
        • Japan Dairy Council
        Guideline for Good-Quality Milk Production.
        https://www.dairy.co.jp/archive/ryoushitu.html
        Date: 2004
        Date accessed: December 20, 2020
        • Japan Policy Council
        For a Growing 21st Century: Stop the Birthrate Decline—Rural Revitalization Strategy.
        http://www.policycouncil.jp/pdf/prop03/prop03.pdf
        Date: 2014
        Date accessed: December 20, 2020
        • Kanamori M.
        • Kondo N.
        Suicide and types of agriculture: A time-series analysis in Japan.
        Suicide Life Threat. Behav. 2020; 50 (31215073): 122-137
        • Karttunen J.P.
        • Rautiainen R.H.
        • Lunner-Kolstrup C.
        Occupational health and safety of Finnish dairy farmers using automatic milking systems.
        Front. Public Health. 2016; 4 (27458580): 147
        • Kato H.
        • Kaichi K.
        • Morita S.
        Changes in the health condition of livestock with reductions of dairy farm workers.
        J. Rakuno Gakuen Univ. 2018; 43: 5-11
        • Kawai K.
        • Kurosawa S.
        • Nagahata H.
        Relationship between milking management practices and milk somatic cell counts on local dairy farms.
        Jpn. J. Anim. Hyg. 2009; 34: 141-147
        • Kearney G.D.
        • Rafferty A.P.
        • Hendricks L.R.
        • Allen D.L.
        • Tutor-Marcom R.
        A cross-sectional study of stressors among farmers in Eastern North Carolina.
        N. C. Med. J. 2014; 75 (25402689): 384-392
        • King M.T.M.
        • Matson R.D.
        • DeVries T.J.
        Connecting farmer mental health with cow health and welfare on dairy farms using robotic milking systems.
        Anim. Welf. 2021; 30: 25-38
        • Kondo K.
        Social determinants of health and rural medicine.
        J. Jpn Assoc. Rural Med. 2019; 67: 636-640
        • Kurihara S.
        • Shimoura S.
        Quality control efforts among Japanese dairy farmers: Findings from national field surveys.
        Hortic. Res. 2008; 62: 77-83
        • Liu Q.
        • He H.
        • Yang J.
        • Feng X.
        • Zhao F.
        • Lyu J.
        Changes in the global burden of depression from 1990 to 2017: Findings from the Global Burden of Disease study.
        J. Psychiatr. Res. 2020; 126 (31439359): 134-140
        • MAFF
        Census of Agriculture and Forestry in Japan: Annual Statistics, 1995 to 2018.
        https://www.maff.go.jp/j/tokei/census/afc/past/stats.html
        Date: 2018
        Date accessed: October 21, 2021
        • MAFF
        Prefectural Food Self-Sufficiency Rate, 2018.
        https://www.maff.go.jp/j/zyukyu/zikyu_ritu/zikyu_10.html
        Date: 2018
        Date accessed: October 21, 2021
        • MAFF
        Milk and Dairy Products Statistics, 2018.
        https://www.maff.go.jp/j/tokei/kouhyou/gyunyu/
        Date: 2018
        Date accessed: October 21, 2021
        • MAFF
        Office for Promotion of Women's Activities, Promotion of Women's Activities in Agriculture.
        • MAFF
        Statistical Survey on Farm Management and Economy, Livestock Production Costs, Hokkaido, 2004–2015.
        • MAFF
        Statistical Survey on Farm Management and Economy, Statistics of Farming Types, 2004–2015.
        https://www.maff.go.jp/j/tokei/kouhyou/noukei/
        Date: 2019
        Date accessed: October 21, 2021
        • MAFF
        Statistics on the Agricultural Labor Force: Population of Agricultural Workers and Number of Core Agricultural Workers, 2020.
        https://www.maff.go.jp/j/tokei/sihyo/data/08.html
        Date: 2020
        Date accessed: October 21, 2021
        • MAFF
        Agriculture, Forestry, and Fisheries in Graphs and Statistics, 2020.12.17update ver.
        http://www.machimura.maff.go.jp/machi/map/01/index.html
        Date: 2020
        Date accessed: October 21, 2021
        • Hokkaido M.A.F.F.
        Annual Report of Statistics on Agriculture, Forestry and Fisheries of Hokkaido 2018-2019.
        • MHLW
        Overview of the estimation of economic benefits of suicide and depression prevention measures (Social losses due to suicide and depression).
        https://www.mhlw.go.jp/stf/houdou/2r9852000000qvsy.html
        Date: 2009
        Date accessed: December 20, 2020
        • MHLW
        Basic survey on wage structure in 2019: Summary of results by industry.
        • Momose Y.
        • Suenaga T.
        • Une H.
        Job satisfaction and mental distress among Japanese farmers.
        J. Rural Med. 2008; 3: 29-33
        • Moorby J.M.
        • Lee M.R.
        • Davies D.R.
        • Kim E.J.
        • Nute G.R.
        • Ellis N.M.
        • Scollan N.D.
        Assessment of dietary ratios of red clover and grass silages on milk production and milk quality in dairy cows.
        J. Dairy Sci. 2009; 92 (19233807): 1148-1160
        • Nishi N.
        • Sugiyama H.
        • Kasagi F.
        • Kodama K.
        • Hayakawa T.
        • Ueda K.
        • Okayama A.
        • Ueshima H.
        Urban-rural difference in stroke mortality from a 19-year cohort study of the Japanese general population: NIPPON DATA80.
        Soc. Sci. Med. 2007; 65 (17493727): 822-832
        • Ohgi T.
        • Kaneko T.
        • Doukoshi A.
        • Kusakari N.
        • Hatta T.
        Effects of milk yield and feeding management on disposal and disease in dairy herds in Hokkaido.
        Bulletin of Hokkaido Prefectural Agricultural Experiment Stations. 2001; 80: 39-44
        • Ohgi T.
        • Shiga E.
        Consider the number of breeding years of a dairy cow—Its facts and determinants.
        Research and Development Center For Dairy Farming, 2001
        • Pearson K.
        On lines and planes of closest fit to systems of points in space.
        Lond. Edinb. Dublin Philos. Mag. J. Sci. 1901; 2: 559-572
        • Sano S.
        • Tanaka K.
        A study of job stressors and stress coping strategies among managers in small and medium-sized enterprises.
        J. Osaka University of Econ. 2012; 63: 161-173
        • Sasano M.
        Quality Control of Raw Milk that Promises Safety and Security to Consumers.
        Research and Development Center for Dairy Farming, 1998: 69-71
        • Schukken Y.H.
        • Wilson D.J.
        • Welcome F.
        • Garrison-Tikofsky L.
        • Gonzalez R.N.
        Monitoring udder health and milk quality using somatic cell counts.
        Vet. Res. 2003; 34 (14556696): 579-596
        • Shima S.
        • Shikano T.
        • Kitamura T.
        • Asai M.
        New self-rating scale for depression.
        J. Clin. Psychiatry. 1985; 27: 717-723
        • Smith K.B.
        • Humphreys J.S.
        • Wilson M.G.
        Addressing the health disadvantage of rural populations: How does epidemiological evidence inform rural health policies and research?.
        Aust. J. Rural Health. 2008; 16 (18318846): 56-66
        • Snow Brand Seed Co
        How milk prices are paid by dairy manufacturers.
        • Tsuru E.
        The Sociology of Farm Women: The Energy of Agriculture Starts with Women.
        Commons, 2007
        • Wallis A.
        • Dollard M.
        Local and global factors in work stress: The Australian dairy farming exemplar.
        Scand. J. Work Environ. Health. 2008; 6: 66-74
        • Yazd S.D.
        • Wheeler S.A.
        • Zuo A.
        Key risk factors affecting farmers' mental health: A systematic review.
        Int. J. Environ. Res. Public Health. 2019; 16 (31810320)4849