J. Dairy Sci.

105:3559–3573
https://doi.org/10.3168/jds.2021-21386
© 2022, The Authors. Published by Elsevier Inc. and Fass Inc. on behalf of the American Dairy Science Association®.
This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).

Cattle farmer psychosocial profiles and their association

with control strategies for bovine viral diarrhea
N. S. Prosser,1* M. J. Green,1 E. Ferguson,2 M. J. Tildesley,3,4 E. M. Hill,3,4 M. J. Keeling,3,4
and J. Kaler1
1
School of Veterinary Medicine and Science, University of Nottingham, Sutton Bonington Campus, Leicestershire, LE12 5RD, United Kingdom
2
School of Psychology, University Park, University of Nottingham, Nottingham, NG7 2RD, United Kingdom
3
Zeeman Institute for Systems Biology & Infectious Disease Epidemiology Research, School of Life Sciences and Mathematics Institute,
University of Warwick, Coventry, CV4 7AL, United Kingdom
4
Joint UNIversities Pandemic and Epidemiological Research (JUNIPER; https:​/​/​maths​.org/​juniper/​)

ABSTRACT motivation (decisions and goals) were more likely to

vaccinate and test, alone or in combination with other
Bovine viral diarrhea (BVD) is endemic in the United controls. Farmers with high psychological proximity
Kingdom and causes major economic losses. Control is (feeling of closeness) to their veterinarian were more
largely voluntary for individual farmers and is likely likely to undertake many prevention strategies in an
to be influenced by psychosocial factors, such as altru- open herd. Farmers with high psychological proximity
ism, trust, and psychological proximity (feeling close) to dairy farmers and low psychological proximity to
to relevant “others,” such as farmers, veterinarians, the beef farmers were more likely to keep their herd closed
government, and their cows. These psychosocial factors and separate and test or vaccinate and test. Farmers
(factors with both psychological and social aspects) who had a lot of trust in other farmers and invested in
are important determinants of how people make deci- them, rather than keeping everything for themselves,
sions related to their own health, many of which have were more likely to be careful introducing new stock
not been studied in the context of infectious disease and test. In conclusion, farmer psychosocial factors were
control by farmers. Farmer psychosocial profiles were associated with strategies for BVD control in UK cattle
investigated using multiple validated measures in an farmers. Psychological proximity to veterinarians was a
observational survey of 475 UK cattle farmers using the novel factor associated with proactive BVD control and
capability, opportunity, motivation-behavior (COM- was more important than the more extensively investi-
B) framework. Farmers were clustered by their BVD gated trust. These findings highlight the importance of
control practices using latent class analysis. Farmers a close veterinarian-farmer relationship and are impor-
were split into 5 BVD control behavior classes, which tant for promoting effective BVD control by farmers,
were tested for associations with the psychosocial and which has implications for successful nationwide BVD
COM-B factors using multinomial logistic regression, control and eradication schemes.
with doing nothing as the baseline class. Farmers who Key words: bovine viral diarrhea, COM-B, farmer
were controlling disease both for themselves and oth- behavior, psychological proximity, psychosocial profiling
ers were more likely to do something to control BVD
(e.g., test, vaccinate). Farmers who did not trust other
INTRODUCTION
farmers, had high psychological capability (knowledge
and understanding of how to control disease), and had Bovine viral diarrhea (BVD) is endemic in the United
high physical opportunity (time and money to control Kingdom and causes major economic losses that result
disease) were more likely to have a closed, separate from poor growth rates and pneumonia, reduced milk
herd and test. Farmers who did not trust other farmers production, reduced fertility, fatalities from mucosal
were also more likely to undertake many prevention disease, and increased susceptibility to other diseases
strategies with an open herd. Farmers with high auto- (Houe, 1999; Weldegebriel et al., 2009). Bovine viral
matic motivation (habits and emotions) and reflective diarrhea transmission occurs primarily through persis-
tently infected (PI) cattle, which are created when the
dam becomes infected early in pregnancy (McClurkin
Received October 6, 2021.
et al., 1984). These calves are immunotolerant to the
Accepted December 5, 2021. BVD virus (Peterhans et al., 2003) and shed virus their
*Corresponding author: Naomi.Prosser@​nottingham​.ac​.uk entire life. Other cattle are transiently infected and act

3559
 Prosser et al.: FARMER PSYCHOSOCIAL PROFILES AND BVD CONTROL 3560

as a weaker source of infection, being infectious for a more intrinsic to a person’s nature than the COM-B
much shorter period. factors.
There are many strategies farmers can adopt to People with altruistic (defined here as seeking to
control BVD, ranging from measures to prevent the benefit others at a personal cost; West et al., 2007;
introduction and transmission of infectious diseases in Bshary and Bergmüller, 2008; Pfattheicher et al., 2022)
general to the BVD specific measures of testing, culling and prosocial (aim for equality) preferences tend to
PI cattle, and vaccination. The decisions farmers make make health decisions that benefit others. In contrast,
within their own herd have implications for national people with more selfish or proself (seek to benefit self)
disease control, with frequent movements between preferences make decisions that benefit themselves. For
cattle holdings and markets in the United Kingdom example, prosocial individuals are more likely to get
(Vernon, 2011) and the potential for local disease trans- vaccinated, unlike proself people who are more likely
mission between neighboring herds (Abernethy et al., to rely on herd immunity (Böhm et al., 2016). People
2011; Graham et al., 2016). There is a potential incen- may also show reactive reluctant altruism, behaving in
tive to free-ride and rely on other people to control a way that benefits others but only because they do
disease (Bauch and Earn, 2004). Where national BVD not trust others to help, which, for example, is impor-
prevalence is high, maintaining freedom from disease in tant in blood donation behavior (Ferguson et al., 2012;
a herd is more costly (Gunn et al., 2005). Bovine viral Ferguson, 2022). Behaviorally, generosity and altruism
diarrhea control is voluntary in England and Wales, can be investigated using dictator games where a deci-
and farmers can engage with voluntary schemes [BVD- sion maker chooses how to split an endowment between
Free (BVDFree, 2021) and Gwaredu BVD (Gwaredu themselves and another person (Forsythe et al., 1994).
BVD, 2019)]. In contrast, BVD testing is mandatory The dictator game can be modified with different recip-
in breeding herds in Scotland and Northern Ireland, ients and different contexts for the original endowment
with movement restrictions on PI cattle to incentivize that is to be split (e.g., it could have been earned, a gift,
control. or lottery winnings) to investigate how altruistic the
The undertaking of any given behavior is influenced decision maker is (Engel, 2011). Altruism can also be
by a person having sufficient capability (e.g., physi- investigated using the social value orientation (SVO)
cal ability, knowledge, and understanding), opportu- slider measure (Murphy et al., 2011) where participants
nity (e.g., physical resources and support from others), make 6 dictator game decisions from a series of set
and motivation (e.g., both reflective decision making responses. Their responses are used to calculate a score
and automatic habits and emotions; Michie et al., of how altruistic they are which can used to categorize
2011, 2014). The capability, opportunity, motivation- participants as competitive (<−12.04, maximizes the
behavior (COM-B) framework of behavior change difference in benefit to self versus benefit to other),
captures these interrelated attributes, encompasses individualistic (−12.04 to 22.45, maximizes benefit to
existing frameworks for behavior in health settings, self), prosocial (22.45 to 57.15, maximizes joint gain or
and has been applied to farmer and veterinary be- minimizes inequality), or altruistic (>57.15, maximizes
havior (Michie et al., 2011; Hardefeldt et al., 2018; benefit to other; Murphy et al., 2011).
Carroll and Groarke, 2019). The COM-B framework People are generally more likely to help those they
has traditionally been used to study predictors of in- feel close to or are psychologically proximal to (Cialdini
dividual behavior change. However, infection control, et al., 1997). High psychological proximity is associ-
be it via vaccination or behavioral controls, requires an ated with increased uptake of behavior to protect
appreciation of the dynamic relationship with people. other people’s health (Tu et al., 2021) and increased
For example, vaccination carries a personal cost, but support for others to change their behavior (Bobak
benefits both the person vaccinated and those not and Raupach, 2018). Psychological proximity can be
vaccinated. Thus, the nonvaccinated can pay no cost measured using the inclusion of other in self (IOS)
and free-ride on others’ decisions to vaccinate (Bauch scale (Aron et al., 1992), where participants select a
and Earn, 2004; Böhm et al., 2016). Therefore, to fully pair of overlapping circles from a scale of increasingly
understand infection control decisions it is necessary overlapping circles that best represents how close they
to include assessment of key mechanisms underlying feel to a specific “other” in question. Trust that oth-
cooperation: generosity, altruism, trust, fairness, and ers will control infectious diseases is another important
proximity (Fehr and Fischbacher, 2003; Nowak, 2006; factor and promotes cooperation with prosocial disease
Rand and Nowak, 2013; Bradley et al., 2018; Dimick et control behavior such as BVD control schemes (Hef-
al., 2018). These psychosocial factors (factors with both fernan et al., 2016; Pletzer et al., 2018). Cattle farmers
psychological and social aspects) come from a separate in the United Kingdom generally have high trust in
theoretical literature to the COM-B framework and are veterinary advice and low trust in government policy.
Journal of Dairy Science Vol. 105 No. 4, 2022
 Prosser et al.: FARMER PSYCHOSOCIAL PROFILES AND BVD CONTROL 3561

They are more likely to follow trusted veterinary advice ity, altruism). The same facilitator conducted all focus
than want to cooperate with government recommenda- groups.
tions (Brennan and Christley, 2013; Fisher, 2013; Bard Focus groups were conducted in February 2020 in
et al., 2019). Behaviorally, trust and trustworthiness 3 geographical areas (Nottinghamshire, Somerset, and
can be investigated with investment games (also known Yorkshire), with 3 to 8 farmers per group. All farm-
as trust games; Berg et al., 1995). In these games the ers had a minimum of 100 dairy cows, and each group
investor chooses how much of an endowment to invest contained both farmers who were part of the national
in another unknown person. The amount invested is BVD eradication program (BVDFree England) and
multiplied (usually tripled) and the investee chooses those who were not. Three of the 4 focus groups also
how much of the multiplied investment to return to the contained both farmers who had regular routine visits
investor. Thus, increasing the amount invested indexes from their veterinarian and those who did not. Farm-
how trusting the investor is that the investee will act ers were recruited as a convenience sample from herds
in a trustworthy manner and return a fair amount on associated with the University of Nottingham (2 focus
the investment. Expectations of trust in others can be groups) and 2 veterinary practices (2 focus groups).
gauged by asking investors to estimate how much they The focus groups took place at the Centre for Dairy
believe an investee will return. Thus, the trust game Science Innovation at the University of Nottingham
also taps concepts of cooperation, and reciprocity (Fer- or at veterinary practices and were recorded and tran-
guson et al., 2020). Modifications can also be made to scribed by an external agency (Penguin Transcription).
this game for the particular context to be studied. Transcripts were checked once against the audio file by
In this paper we investigate psychosocial profiles of the author (NP). Theoretical thematic analysis (Braun
UK cattle farmers, many of which have not been inves- and Clarke, 2006) of all transcripts was conducted to
tigated in farmers to date, to evaluate how individual identify and explore the psychosocial constructs that
profiles and factors from the COM-B framework of be- should be included on the broader farmer survey.
havior change, are associated with the farmers’ strategy
to control BVD in their herd. Survey Design

MATERIALS AND METHODS A survey was designed to cover a wide spectrum of

relevant other-regarding preferences. These included
Ethical approval was obtained from the University altruism (West et al., 2007; Bshary and Bergmüller,
of Nottingham Research Ethics Committee for both 2008; Pfattheicher et al., 2022), reactive reluctant al-
the focus groups and the survey before commencement truism (helping because of lack of trust that others will
of the study (reference number: 2789 190711, granted: help; Ferguson et al., 2012; Ferguson, 2022), and trust
June 22, 2019). and distrust (McKnight and Chervany, 2001). Altruism
was assessed using the SVO slider measure questions in
Focus Group Design, Recruitment, and Analysis which farmers chose to allocate money between them-
selves and another unknown farmer (Murphy et al.,
A series of focus groups (n = 4) was conducted to in- 2011). This task was incentivized; 10 farmers were se-
form the development of a farmer survey of psychosocial lected at random to receive a financial reward for one of
characteristics and infectious disease control. A focus their decisions, paid to themselves and another random
group guide was designed to investigate altruism, trust, survey respondent; payment based on one decision is
psychological proximity, and the COM-B framework of commonly used for this type of economic decision and
behavior change (Michie et al., 2011) in the context of has the advantage of reducing hedging (Charness et al.,
BVD control in a 1-h discussion. The facilitator used 2016). Altruism was also investigated using a dictator
the questions and prompts in the focus guide to start game where farmers could share £700 ($877.80) hypo-
the discussion and ensure that the farmers discussed all thetical lottery money between themselves, an unknown
areas of interest, but otherwise allowed the farmers to farmer, a known farmer, an unknown veterinarian, a
talk freely. During the focus group, participants were known veterinarian, and a stranger. Hypothetical pay
provided with a printout of the IOS scale, a measure does not alter dictator game decisions (Engel, 2011).
of psychological proximity (Aron et al., 1992; Mashek Trust was investigated using an investment game (Berg
et al., 2007), as a discussion aid (Supplemental Figure et al., 1995) where £50 hypothetical lottery winnings
S1; https:​/​/​rdmc​.nottingham​.ac​.uk/​handle/​internal/​ could be invested in another unknown farmer. Invest-
9483). The question guide was modified slightly be- ments were tripled and the farmers were asked how
tween the first and subsequent focus groups to improve much they would expect the other farmer to return. Al-
probing on other-regarding preferences (e.g., prosocial- truism, reluctant altruism, trust, and distrust were also
Journal of Dairy Science Vol. 105 No. 4, 2022
 Prosser et al.: FARMER PSYCHOSOCIAL PROFILES AND BVD CONTROL 3562

investigated by Likert-scale questions created by the dom selection for the farmers who were rewarded with
authors investigating trust and distrust in farmers, vet- a payment based on their SVO decisions was conducted
erinarians, government, strangers, the National Farm- in R statistical software (v3.6.2; R Core Team, 2019).
ers’ Union, and the farming press (section 4, question
4 in Supplemental File S1; https:​/​/​rdmc​.nottingham​.ac​ Data Entry and Analysis
.uk/​handle/​internal/​9483).
We assessed factors associated with other-regarding Data entry for the postal surveys was conducted by
preferences such as psychological proximity (Aron et an outside agency (Wyman Dillon Ltd.), except for
al., 1992; Mashek et al., 2007), general motivation for the final 8 late returns which were conducted by the
behavior using the COM-B framework (Michie et al., author (NP). All data analysis was conducted using R
2014), and anxiety about BVD. Psychological proxim- statistical software (v3.6.2; R Core Team, 2019) and
ity with farmers, veterinarians, the government, the Na- each analysis used only the complete responses for the
tional Farmers’ Union, and their cows was investigated relevant survey questions.
using the IOS scale (Aron et al., 1992; Mashek et al.,
2007). Each set of increasingly overlapping circles was Factor Analyses
given a score of 1 to 7, with higher numbers for increased
overlap which represented increased psychological prox- To identify latent groupings and reduce the number
imity (Supplemental Figure S1). The COM-B questions of variables for analysis, factor analysis (Thurstone,
were based on a published question guide (Michie et al., 1947) was conducted on 32 Likert-scale items regard-
2014), with additional relevant questions from other ing reluctant altruism, altruistic or proself preferences,
published uses of the guide (Barker et al., 2016; Taylor trust, and distrust. The factor analysis was conducted
et al., 2016; Bobak and Raupach, 2018). The COM-B using the psych package (v2.0.8) with maximum likeli-
questions were used to explore psychological capability hood and oblimin rotation (Revelle, 2020). Likert-scale
(knowledge and understanding), physical opportunity questions were converted to numeric (1–5 for ascend-
(time and money), social opportunity (support from ing strength of agreement). The number of factors
others), automatic motivation (habits and emotions), was chosen based on parallel analysis, fit statistics
and reflective motivation (plans and goals). Physical (Tucker-Lewis index ≥0.9 and root mean squared error
capability (ability) was not considered relevant so was of approximation ≤0.06; Hu and Bentler, 1999), and a
not investigated in the survey. We considered that a minimum of 2 variables loaded to each factor. Loadings
lack of physical capability would preclude farmers from ≥0.3 were considered to define a factor. Items that were
keeping cattle; therefore, it was not applicable to this cross-loaded on 2 factors were retained in the model
group of people. Anxiety about BVD was assessed on a if their omission did not change the model fit. Factor
5-point scale of how anxious farmers felt about a BVD scores for each respondent were calculated from every
breakdown on their farm. Farmers were also asked statement in the factor analysis, weighted by its load-
questions to gather demographic information and to ing (regardless of size of loading). Cronbach’s α was
capture their current implementation of BVD control calculated for each factor to assess internal reliability.
strategies. If Cronbach’s α was <0.7 and could be improved with
The survey was tested by members of the research omission of an item, the item was omitted.
group and in a pilot study conducted using a conve- The COM-B items were grouped by their overarching
nience sample of 8 cattle farmers. The final survey in factor (psychological capability, physical opportunity,
full is provided in Supplemental File S1. social opportunity, automatic motivation, and reflec-
tive motivation). The Likert-scale responses were also
Survey Dissemination converted to a numeric response (1–5 for ascending
strength of agreement). Factor scores were created
The survey was open from July 13 to October 5, by taking the mean score of the items in that factor,
2020. The survey link was emailed to 10,560 British with an item’s scale inverted if necessary (Supplemen-
dairy and English beef levy payers by the Agriculture tal Table S1; https:​/​/​rdmc​.nottingham​.ac​.uk/​handle/​
and Horticulture Development Board, and various internal/​9483).
cattle interest organizations also promoted the survey
via magazine articles, e-newsletters, social media posts, Latent Class Analysis
website posts, and emails to UK cattle farmers. A hard-
copy was also posted to a random sample of 2,000 of Two latent class analyses were conducted to identify
the dairy and 2,000 of the beef levy payers. The ran- clusters of farmers from their responses:

Journal of Dairy Science Vol. 105 No. 4, 2022

 Prosser et al.: FARMER PSYCHOSOCIAL PROFILES AND BVD CONTROL 3563

(1) Three items from the economic games were res- Northern Ireland, and English counties north of and
caled to proportions for comparative purposes: including Cheshire, Derbyshire, Nottinghamshire, and
the proportion of £50 that each farmer invested, Lincolnshire) and southern (Wales and all remaining
the proportion of tripled investment each farmer English counties) areas to test in the model. Addition-
expected to be returned in the investment game, ally, the southwest (Gloucestershire, Wiltshire, Dorset,
and the proportion of £700 each farmer gave and more southwesterly counties) were separated out
away in the dictator game. and the variable retested in the model (north, south,
(2) The BVD control behaviors of the farmers: buy southwest). Model fit was evaluated by predicting BVD
only from BVD-free herds, closed herd, disinfec- behavior class from both the full and 10 × 10-fold
tion for people entering the farm, separation cross-validated models and comparing the proportion
from neighboring stock, isolate or test new cattle, of times the correct class was predicted, and with a
vaccinate, blood or tissue test, milk test, cull PI, Hosmer-Lemeshow goodness-of-fit test (Fagerland and
and isolate sick animals. Hosmer, 2012).
Conditional autoregressive models (Lee, 2013) were
Latent class analysis was conducted using the mclust created to investigate potential spatial confounding of
(v5.4.6) package (Fraley and Raftery, 2002; Scrucca the psychosocial factors at the county level. No con-
et al., 2016). Models were selected as the spherical or founding was identified and the methods and results for
diagonal mixture model with the highest Bayesian in- these are presented in Supplemental File S2 (https:​/​/​
formation criterion (BIC) where at least 5% of farmers rdmc​.nottingham​.ac​.uk/​handle/​internal/​9483).
were in each class and there was good delineation of the
classes (normalized entropy >0.7).
RESULTS

Multinomial Logistic Regression Descriptive Statistics of Survey Respondents

To investigate associations between farmer BVD con- A total of 291 online survey responses were received
trol behaviors and psychosocial factors, the BVD latent from July 16 to October 5, 2020, which was 50.4%
classes were evaluated as a multinomial outcome in a (/577) of started surveys. A further 184 surveys were
logistic regression model with explanatory covariates: returned by post (response rate of 4.6%), making a to-
economic games latent class, altruism and trust fac- tal of 475 survey responses. The mean pay-out to the 20
tors, psychological proximity to others, each COM-B prize-winning farmers was £74.25 (range £15–£100).
factor (psychological capability, physical opportunity, Most respondents were in England (73%), with 14%
social opportunity, automatic motivation, and reflec- in Scotland, 7% in Wales, and 2% in Northern Ireland.
tive motivation), and anxiety about a BVD breakdown. Most farmers were in their 50s and 60s (30% and 25%,
The models were built with the nnet (v7.3.12) package respectively), with 7% under 30, 10% in their 30s, 16%
(Venables and Ripley, 2002). in their 40s, and 11% over 70. Seventy-six percent of
Each independent variable was initially tested in a farmers had beef cattle, with a median of 70 animals
univariable model and initially all significant variables over 6 mo old (range 0–850), and 39% had dairy cattle,
(P < 0.05) were selected to test in a multivariable with a median of 180 adult cows (range 2–1,309).
model. Terms that were not significant in the multilevel
model were removed from the model and all terms were Farmer SVO and Levels of Trust and Distrust
retested in the final model. Spearman rank correlations
were calculated for all variables considered for inclu- Most farmers were categorized by their SVO slider
sion in the multivariable model. Variables that were measure responses as prosocial (75.4%), 16.6% were
correlated (≥0.3) with a variable in the final model individualistic, and very few were altruistic (1.3%) or
were tested in the model in place of their correlated competitive (0.4%).
variable. Models where correlated terms were signifi- Veterinarians were the most trusted group, least dis-
cant when substituted in the final model are presented trusted group, and farmers also felt the most respected
as alternative models. Demographic data were not in- by them. Eighty-one percent of farmers trusted their
cluded in the multivariable model because these were veterinarian compared with the National Farmers’
all categorical and there were insufficient respondents Union (57%), dairy farmers (47%), beef farmers (38%),
to avoid small and empty contingency table cells in and governmental organizations (19%). More farmers
the model. However, each demographic variable was agreed that they felt respected by their veterinarian
tested in the final model to check for confounding. The (85%), with 67% agreeing that they felt respected by
United Kingdom was split into northern (Scotland, the veterinary profession, whereas 49% and 10% felt
Journal of Dairy Science Vol. 105 No. 4, 2022
 Prosser et al.: FARMER PSYCHOSOCIAL PROFILES AND BVD CONTROL 3564
Table 1. The items loaded (>0.3) onto each of the 7 factors from a factor analysis of 436 complete responses to Likert-scale items on reluctant
altruism, altruistic or proself preferences, trust, and distrust in a survey of UK cattle farmers

Factor1 Item2 Loading

Trust in National Farmers’ Union I feel respected by the National Farmers’ Union. 0.99
(α = 0.91) I trust the National Farmers’ Union. 0.82
Reluctant altruism I vaccinate my cows because I cannot trust other farmers to vaccinate theirs. 0.94
(α = 0.87) I vaccinate my cows to protect my herd and those around me, because other farmers 0.80
will not vaccinate.
Trust in veterinarians I trust my vet’s advice about infectious disease control in my herd. 0.88
(α = 0.87) My vet would always tell me the truth even if it was not what I wanted to hear. 0.81
I trust vets. 0.61
Farmers receive high quality veterinary advice from the veterinary profession. 0.59
I feel respected by my vet. 0.58
I feel respected by the veterinary profession. 0.49
Trust in farmers I trust my neighbors to be controlling infectious diseases in their herds. 0.78
(α = 0.79) I trust other farmers nationally to be controlling infectious diseases in their herd. 0.70
I trust beef farmers. 0.65
I trust dairy farmers. 0.50
I trust other farmers I meet for the first time. 0.41
Controlling disease for self and others I control infectious disease because I take pride in having a healthy herd. 0.86
(α = 0.81) I control infectious disease to protect my reputation for having healthy cattle. 0.70
Controlling infectious disease in the UK will have benefits for every farmer. 0.69
I control infectious disease to do my bit for national disease control. 0.68
I control infectious disease in my cattle to protect my own herd. 0.51
I control infectious disease in my cattle to protect other farmers’ herds. 0.43
Trust in government I trust governmental judgments about how to control infectious diseases in cattle. 0.78
(α = 0.76) I feel respected by the government. 0.76
I trust governmental organizations. 0.65
When dealing with the government it is better to be careful before you trust them. −0.44
General distrust When dealing with strangers it is better to be careful before you trust them. 0.48
(α = 0.57) When dealing with farmers it is better to be careful before you trust them. 0.47
When dealing with vets it is better to be careful before you trust them. 0.40
In general, one can trust people. −0.43
Items that did not load on any factor I only control infectious disease when other farmers are also taking steps to control —
disease.
I control infectious disease to stay ahead of other farmers. —
Infectious disease information in the farming press is trustworthy. —
1
α = Cronbach’s α.
2
vet = veterinarian.

respected by the National Farmers’ Union and gov- A 7-factor solution gave the best fit in a factor analy-
ernment, respectively. Similarly, only 31% of farmers sis of 436 complete responses to the other-regarding
thought it was better to be careful before you trust vet- preferences, reluctant altruism, trust, and distrust
erinarians, rising to 56% for farmers, 63% for govern- Likert-scale measures (Tucker-Lewis index = 0.88, root
ment, and 76% for strangers. Only 17% trusted farmers mean squared error of approximation = 0.06; Table 1).
they met for the first time. “I trust other farmers I meet for the first time” loaded
Trust in veterinarians was even higher when it was on both “trust in farmers” and “general distrust” fac-
specifically for veterinary services and infectious disease tors (0.41 and −0.35, respectively) and was retained in
control compared with general trust: 90% trusted their the model because its omission did not alter the model.
veterinarian’s advice, 90% agreed that their veterinar- Factor loadings are presented in Table 1. Cronbach’s
ian would always tell them the truth even if it was not α was >0.7 for all factors except “general distrust”;
what they wanted to hear, and 80% agreed that farm- however, this was not improved by omission of any of
ers received high-quality advice from the veterinary its items (Table 1).
profession. There was less trust in other stakeholders,
with 43% agreeing that infectious disease information Farmer Behaviors in the Economic Games
in the farming press was trustworthy, 35% and 30%
trusting their neighbors and other farmers nationally, Seventy-four percent of farmers made an investment
respectively, to control infectious diseases, and only in the trust game. The mean proportion invested by
16% of farmers trusting governmental judgments about the investors was 0.53 of £50 (range 0.02–1.00) and
disease control. the mean proportion of the tripled investment that the

Journal of Dairy Science Vol. 105 No. 4, 2022

 Prosser et al.: FARMER PSYCHOSOCIAL PROFILES AND BVD CONTROL 3565

Figure 1. The median responses and interquartile range of the 4 latent classes of the behaviors of 417 farmers in the economic games. Invest
= proportion of £50 ($62.70) invested in an unknown farmer in the investment game; return = proportion of the tripled investment expected
back from the unknown farmer; give = proportion of £700 given away to others in the dictator game. The latent classes are as follows: 1 =
generous self-oriented mutual benefit: the most generous class that invests some and expects other farmers to be fair; 2 = Homo economicus
(selfish): keeps everything and does not trust other farmers to be fair; 3 = mutually beneficial joint maximizer: risk taking in investing everything
in unknown farmers who they trust to be fair; 4 = self-oriented mutual benefit: cautiously invests a small amount, keeping most of the possible
investment from the other farmer who they do not trust to be fair as much as the other 2 investing classes.

investors expected the other farmer to return was 0.42 and kept everything (30% of farmers); these farmers
(range 0.00–1.00). could be classed as selfish profit maximizers who do not
Forty-seven percent of farmers gave some money trust other farmers to be fair. The third largest class,
in the dictator game, with both the total generosity “mutually beneficial joint maximizer,” invested every-
and its distribution between the recipients varying by thing, expected an equal split of the investment back,
farmer. For the farmers who gave money, the mean pro- and gave some away (19% of farmers). This investment
portion given away was 0.39 of £700 (range 0.01–1.00). strategy maximizes the number of resources available
The farmers were most generous to neighboring farmers to farmers in general, and therefore farming. This class
with a mean gift of 0.16 of the lottery money, the local of farmer would end up with the most total resources
veterinarian received 0.12, the unknown farmer 0.05, if the investment recipient returns the proportion of in-
the stranger 0.04, and the unknown veterinarian 0.02. vestment that the farmer expects. This strategy reveals
Farmers split into 4 latent classes to describe their a high level of trust in other farmers and a willingness to
investment and generosity decisions (Figure 1), using risk investing everything in other farmers. The final and
the complete responses from 417 farmers. An ellip- smallest class, “self-oriented mutual benefit,” invested
soidal, equal volume, and equal shape mixture model some, expected a less than equal split of the investment
had the best BIC (BIC = 412.67, log-likelihood = back, and gave very little away (16% of farmers). This
296.83, normalized entropy = 0.98). The 4 classes are class is less altruistic than the “generous self-oriented
described in detail in Supplemental File S3 (https:​/​/​ mutual benefit” class and invested less. These farmers
rdmc​.nottingham​.ac​.uk/​handle/​internal/​9483). Briefly, are cautiously investing only a small amount to farmers
farmers in the first and largest class, “generous self- who they do not trust to be very fair.
oriented mutual benefit,” invested half, expected an
equal split of the investment back, and were generous Farmer Psychological Proximity to Others
(35% of farmers). This class was the most altruistic in
the dictator game, but not willing to invest everything The farmers felt closer to their cows (mean score of
in farmers that they did not completely trust, so kept 6.0, Supplemental Figure S1) than any of the groups of
half of the possible investment. This investment game people and were closer to known groups than unknown
strategy means that farmers benefit overall but the groups of people. Similar to their level of trust, the
investing farmer benefits the most. The second largest farmers felt closer to their veterinarian (4.8) than to
class, “Homo economicus (selfish),” invested nothing neighboring farmers (3.9), the veterinary community

Journal of Dairy Science Vol. 105 No. 4, 2022

 Prosser et al.: FARMER PSYCHOSOCIAL PROFILES AND BVD CONTROL 3566

(3.5), dairy farmers (3.4), beef farmers (3.4), the farm- likely to be undertaking many prevention strategies in
ing community (3.3), the National Farmers’ Union an open herd. Farmers with high psychological proxim-
(3.0), and the government (2.0). There was a range of ity to dairy farmers and low psychological proximity to
scores of 1 to 7 for all groups. beef farmers were more likely to keep their herd closed
and separate and test or vaccinate and test than do
BVD Control Methods Used by Farmers nothing. Finally, farmers who were in the “mutually
beneficial joint maximizers” latent class rather than
Farmers split into 9 latent classes to describe their selfish were more likely to be careful introducing new
BVD control decisions using the complete responses stock and testing rather than doing nothing.
from 467 farmers; detailed descriptions of each class There was no evidence of confounding of the demo-
are given in Table 2. A spherical, equal volume mix- graphic variables and little evidence of spatial autocor-
ture model had the best BIC (BIC = −5,269.64, log- relation (Supplemental File S2). There was no evidence
likelihood = −2,330.58, entropy = 0.89). Farmers were of poor model fit in a Hosmer-Lemeshow goodness-of-fit
fairly evenly split across all 9 classes: “doing nothing” test (χ2 = 25.9, df = 32, P = 0.768) or when evaluating
(12%), “closed, separate and testing” (15%), “vaccinat- cross-validated predictions.
ing” (17%), “vaccinating, testing and culling” (8%),
“careful introducing new stock and testing” (9%), DISCUSSION
“careful introducing new stock, separate, and testing”
(7%), “careful introducing new stock, separate, test- This study aimed to investigate farmer psychosocial
ing, and disinfecting people” (8%), “careful introducing factors and their association with the undertaking of
new stock, separate, vaccinating, and testing” (14%), on-farm BVD control measures. A key strength of the
“careful introducing new stock, separate, vaccinating, study was that we used established psychosocial mea-
testing, and disinfecting people” (11%). sures and theory to underpin our survey questions and
analysis. The major findings were that psychological
Multinomial Logistic Regression proximity to veterinarians, trust in farmers, automatic
of Disease Prevention and reflective motivation, psychological capability,
and physical opportunity were important for farmer
Latent classes that contained similar BVD manage- behavior regarding BVD control strategies; these are
ment strategies were merged to reduce the number of discussed in turn below.
classes in the multinomial model, resulting in 5 classes Psychological proximity to veterinarians is a novel
(Table 3). Briefly, these classes were (1) doing noth- psychosocial construct of importance for cattle infec-
ing, (2) closed herd, separated from neighboring stock tious disease control. Farmers who felt close to their
and testing, (3) vaccinating and testing, (4) careful veterinarian were more likely to do more to prevent and
introducing new stock and testing, and (5) undertaking control BVD in an open herd. This is a novel field of
many prevention strategies with an open herd. Univari- application for the IOS scale and there is only limited
able model results examining demographic variables research into psychological proximity in human health
and psychological constructs are provided in Supple- behavior (Bobak and Raupach, 2018; Tu et al., 2021).
mental Tables S2 and S3 (https:​/​/​rdmc​.nottingham​ The IOS scale correlates with both feeling close and
.ac​.uk/​handle/​internal/​9483), and the multivariable behaving close (Aron et al., 1992), which could explain
model results are presented in Table 4. why farmers who had higher psychological proximity to
In the multivariable model of 380 farmers, farmers their veterinarian were more likely to do more to prevent
who “controlled disease for themselves and others” and control BVD if they had an open herd, behaviors
were more likely to be in any class for BVD control that tend to align with veterinary advice. Trust is part
than doing nothing. Farmers who did not trust other of the interpretation that respondents give to the IOS
farmers, had high psychological capability, and had scale and the two are correlated (Kong, 2018; Kleinert
high physical opportunity were more likely to have a et al., 2020). Therefore, farmers who felt closer to their
closed, separate herd and be testing than doing noth- veterinarian also had greater trust in veterinarians
ing. Farmers who did not trust other farmers were also (correlation coefficient = 0.52). However, psychological
more likely to be undertaking many prevention strate- proximity includes aspects other than just trust: behav-
gies with an open herd. Farmers with high automatic ioral closeness, connection with the other, independence
and reflective motivation were more likely to be in the from the other, and similarities with the other, which
classes that were vaccinating and testing, alone or in is also part of how respondents interpret the IOS scale
combination with other controls. Farmers with high (Aron et al., 1992). Trust in veterinarians is commonly
psychological proximity to their veterinarian were more found as an important factor in cattle farmer infectious
Journal of Dairy Science Vol. 105 No. 4, 2022
 Table 2. The percentage of 467 farmers in each of the 9 latent classes describing the farmer’s approach to control of bovine viral diarrhea (BVD) and a description of the control
measures undertaken by each class (with the percentage of farmers using a control measure in parentheses)1

Typical practice Frequent practice Atypical practice

Class Percent (>60% of farmers) (40–60% of farmers) (<40% of farmers)
Doing nothing 12 Isolate or test new cattle (34%)
Closed (25%)
Milk test (13%)
Buy only from BVD-free herds (8%)
Separate from neighboring stock (8%)
Blood or tissue test (8%)
Cull PI (7%)
Disinfect people (5%)
Vaccinate (4%)
Closed, separate, and testing 15 Closed (99%) Disinfect people (58%) Cull PI (9%)
Separate (84%) Milk test (57%)

Journal of Dairy Science Vol. 105 No. 4, 2022

Blood or tissue test (77%) Vaccinate (51%)
Vaccinating 17 Vaccinate (98%) Blood or tissue test (55%) Closed (33%)
Milk test (42%) Isolate or test new cattle (31%)
Buy only from BVD-free herds (17%)
Disinfect people (6%)
Separate from neighboring stock (5%)
Cull PI (2%)
Vaccinating, testing, and culling 8 Cull PI (98%) Only buy from BVD-free herds (33%)
Blood or tissue test (88%) Separate from neighboring stock (22%)
Vaccinate (86%) Closed (17%)
Milk test (66%) Disinfect people (13%)
Isolate or test new cattle (63%)
Careful introducing new stock and testing 9 Isolate or test new cattle (81%) Closed (15%)
Blood or tissue test (88%) Milk test (7%)
Buy only from BVD-free herds (70%) Vaccinate (6%)
Disinfect people (3%)
Separate from neighboring stock (1%)
Careful introducing new stock, separate, 7 Separate from neighboring stock (99%) Closed (23%)
and testing Blood or tissue test (87%) Cull PI (12%)
Isolate or test new cattle (84%) Milk test (2%)
Buy only from BVD-free herds (62%) Vaccinate (1%)
Disinfect people (1%)
Careful introducing new stock, separate, 8 Isolate or test new cattle (100%) Closed (14%)
testing, and disinfecting people Disinfect people (99%) Milk test (9%)
Buy from BVD-free herds (89%) Cull PI (9%)
Blood or tissue test (86%) Vaccinate (2%)
Prosser et al.: FARMER PSYCHOSOCIAL PROFILES AND BVD CONTROL

Separate from neighboring stock (64%)

Careful introducing new stock, separate, 14 Vaccinate (99%) Cull PI (17%)
vaccinating, and testing Test or isolate new stock (92%) Milk test (16%)
Blood or tissue test (89%) Closed (8%)
Buy from BVD-free herds (88%) Disinfect people (1%)
Separate from neighboring stock (84%)
Careful introducing new stock, separate, 9 Disinfect people (99%) Cull PI (26%)
vaccinating, testing, and disinfecting Vaccinate (98%) Milk test (20%)
people Test or isolate new stock (95%)
Blood or tissue test (92%)
Separate from neighboring stock (87%)
Buy only from BVD-free herds (86%)
1
PI = persistently infected.
3567
 Table 3. The percentage of 467 farmers in each of the 5 latent classes describing the farmer’s approach to control of bovine viral diarrhea (BVD) and a description of the control
measures undertaken by each class (with the percent of farmers using a control measure in parentheses)1

Typical practice Frequent practice Atypical practice

Class Percent (>60% of farmers) (40–60% of farmers) (<40% of farmers)

Journal of Dairy Science Vol. 105 No. 4, 2022

Doing nothing 12 Isolate or test new cattle (34%)
Closed (25%)
Milk test (13%)
Buy only from BVD-free herds (8%)
Separate from neighboring stock (8%)
Blood or tissue test (8%)
Cull PI (7%)
Disinfect people (5%)
Vaccinate (4%)
Closed, separate and testing 15 Closed (99%) Disinfect people (58%) Cull PI (9%)
Separate from neighboring stock (84%) Milk test (57%)
Blood or tissue test (77%) Vaccinate (51%)
Vaccinating and testing 25 Vaccinate (94%) Milk test (50%) Cull PI (33%)
Blood or tissue test (66%) Isolate or test new cattle (41%) Closed (28%)
Buy only from BVD-free herds (22%)
Separate from neighboring stock (10%)
Disinfect people (8%)
Careful introducing new stock and testing 16 Blood or tissue test (88%) Separate from neighboring stock Closed (19%)
Isolate or test new cattle (82%) (44%) Cull PI (15%)
Buy only from BVD-free herds (67%) Milk test (5%)
Vaccinate (4%)
Disinfect people (2%)
Careful introducing new stock, separate, 31 Test or isolate new cattle (95%) Disinfect people (55%) Cull PI (18%)
vaccinating and testing Blood or tissue test (89%) Milk test (15%)
Buy only from BVD-free herds (88%) Closed (11%)
Separate from neighboring stock (80%)
Prosser et al.: FARMER PSYCHOSOCIAL PROFILES AND BVD CONTROL

Vaccinate (74%)
1
PI = persistently infected.
3568
 Table 4. The results of a multivariable multinomial model of 380 UK cattle farmers in 5 bovine viral diarrhea (BVD) control classes explained by farmer psychosocial attitudes
and capability, opportunity, motivation-behavior (COM-B) factors1

Careful introducing
Closed, separate, Careful introducing new stock, separate,
and testing Vaccinating and testing new stock and testing vaccinating, and testing

Journal of Dairy Science Vol. 105 No. 4, 2022

Reference BVD behavior class
= doing nothing OR 95% CI P-value OR 95% CI P-value OR 95% CI P-value OR 95% CI P-value
Generous self-oriented mutual 1.17 0.38–3.58 0.787 1.22 0.47–3.17 0.683 0.83 0.30–2.31 0.728 1.77 0.68–4.56 0.239
benefit
Ref = Homo economicus (selfish)
Mutually beneficial joint maximizer 2.75 0.54–14.09 0.226 2.16 0.46–10.15 0.331 4.72 1.05–21.15 0.043 3.76 0.83–16.97 0.085
Ref = Homo economicus (selfish)
Self-oriented, mutual benefit 2.28 0.49–10.63 0.293 2.34 0.61–9.01 0.217 0.94 0.21–4.22 0.930 3.64 0.97–13.63 0.055
Ref = Homo economicus (selfish)
Controlling disease for self and 2.73 1.54–4.81 0.001 2.03 1.31–3.14 0.001 1.48 1.00–2.20 0.050 2.55 1.65–3.92 <0.001
others
Reflective motivation {0.57} 2.71 0.96–7.67 0.061 3.27 1.32–8.09 0.011 2.03 0.79–5.18 0.141 4.89 2.00–11.98 0.001
Automatic motivation {0.50} 1.93 0.74–5.00 0.177 2.77 1.16–6.61 0.021 2.02 0.83–4.94 0.124 3.95 1.69–9.26 0.002
Trust in farmers 0.51 0.28–0.91 0.024 0.79 0.46–1.36 0.392 0.84 0.49–1.45 0.533 0.52 0.31–0.89 0.016
Psychological proximity to your 1.28 0.94–1.75 0.118 1.25 0.95–1.65 0.114 1.06 0.79–1.40 0.709 1.32 1.01–1.74 0.041
veterinarian
Psychological proximity to dairy 2.26 1.39–3.67 0.001 2.03 1.30–3.17 0.002 0.91 0.58–1.43 0.688 1.08 0.71–1.63 0.727
farmers
Psychological proximity to beef 0.37 0.22–0.62 <0.001 0.45 0.28–0.71 0.001 1.05 0.66–1.68 0.834 0.84 0.54–1.30 0.432
farmers
Psychological capability 5.65 1.99–16.05 0.001 1.50 0.61–3.70 0.383 1.30 0.52–3.26 0.572 2.08 0.86–5.00 0.102
Reflective motivation {0.55} 2.32 0.81–6.68 0.117 1.96 0.79–4.85 0.146 1.50 0.61–3.73 0.378 2.85 1.16–6.98 0.022
Automatic motivation {0.49} 1.65 0.61–4.47 0.321 1.82 0.75–4.40 0.182 1.51 0.63–3.65 0.357 2.55 1.07–6.03 0.034
Physical opportunity {0.32} 1.86 1.01–3.44 0.046 1.18 0.69–2.03 0.540 1.33 0.75–2.34 0.324 1.33 0.78–2.27 0.287
Prosser et al.: FARMER PSYCHOSOCIAL PROFILES AND BVD CONTROL

1
Variables correlated with variables in the model (correlation coefficient in brackets {}) and that would be included (P < 0.05) with the omission of the correlated factor are indented
in the table and presented as alternatives below the correlated variable. Ref = reference economic games latent class; all other categories are on a continuous scale. OR = odds ratio.
3569
 Prosser et al.: FARMER PSYCHOSOCIAL PROFILES AND BVD CONTROL 3570

disease control behavior in the literature (e.g., Brennan therefore, although lack of trust in farmers is positive
and Christley, 2013; Hernández-Jover et al., 2016; Bard in terms of farmers taking responsibility for protecting
et al., 2019); however, here psychological proximity was their own herd from BVD, the tendency of farmers to
more important. Veterinarians and farmers frame bi- think only of their own herd has implications for being
osecurity in different ways, hindering the veterinarian- able to achieve national disease control via voluntary
farmer relationship (Shortall et al., 2016). Therefore, and cohesive farmer action.
veterinarians have a role in increasing the psychological In terms of general other-regarding preferences,
proximity that farmers feel with them and encouraging 75.4% of farmers were categorized as prosocial, with
greater uptake of veterinary advice by paying attention almost everyone else classed as individualistic. This
to the broader aspects of psychological proximity than distribution of SVO categories was similar to other
only trust, such as taking up a farmer perspective on studies but with a higher proportion of prosocials than
disease control. the typical 65% (Murphy and Ackermann, 2014). Pro-
In contrast to the relationship with veterinarians, socials can be further divided into those who wish to
trust in farmers was more important than psychologi- minimize inequality between themselves and someone
cal proximity for BVD control, with a lack of trust in else and those who wish to maximize joint gain. Differ-
farmers associated with farmers either maintaining a entiating between these groups requires additional SVO
closed herd separated from neighboring stock or using slider measure questions (Murphy et al., 2011), which
many methods of control if they had an open herd. were not included in the survey to maintain brevity
Farmers often ensure that they buy animals from other and achieve a good response rate. Further research into
farmers that they trust (Hernández-Jover et al., 2016), how prosocial UK farmers subdivide and any associa-
and maintaining a closed, separate herd offers even tions with infectious disease control behavior would be
more protection against disease that may be transmit- worthwhile. When exploring broader other-regarding
ted from farmers that are not trusted. Lack of trust in preferences (i.e., altruism, prosociality, and trust) in
other farmers has previously been reported as a barrier the dictator and trust games we identified 4 classes
to biosecurity uptake, with farmers unwilling to con- of other-regarding preferences for farmers. Therefore,
tribute to collective action that they do not trust other farmers are heterogeneous in terms of the other-re-
farmers to engage in (Heffernan et al., 2008; Shortall et garding strategies and these differences need to be ac-
al., 2016). This leads them to support greater regula- counted for to understand how farmers’ other-regarding
tion by government to ensure that all farmers play their preferences influence behavior.
part in controlling disease (Heffernan et al., 2016). This Farmers with high motivation to control infectious
opinion was also voiced by some of the farmers in the diseases were more likely to use vaccination, especially
focus groups in this study in the context of national with other preventive measures. Both reflective (goals
BVD eradication. In contrast, farmers who were pre- and decision making) and automatic (habits and emo-
pared to invest everything in another unknown farmer tions) motivation were associated with vaccine use, with
in the economic games with a high expectation that the a slightly higher effect from reflective motivation. As-
other farmer will be fair in return, the “mutually ben- pects of reflective motivation have been well researched
eficial joint maximizers” class, were more likely to take in infectious disease control and farmers are more likely
care introducing new stock and test their cattle, than to control disease in their herd if they take responsi-
do nothing. High investment in the investment game bility for disease control, want to see the benefits of
indicates both a high level of trust in the other farmer controlling infectious diseases, or have goals to reduce
and a willingness to take risks (Chetty et al., 2021). or remove disease from their herd (Ellis-Iversen et al.,
Maximizing the growth of the investment also leads 2010; Azbel-Jackson et al., 2018; Robinson, 2020). The
to the greatest resource to the farming community, emotional aspects of automatic motivation are known to
regardless of whether the other farmer returns any of affect farmer behavior (O’Kane et al., 2017). Infectious
the investment to the donor farmer. These community- disease is frequently an emotive subject for farmers and
minded, trusting farmers were more likely to rely on worry often leads farmers to take preventive action to
testing and the status of the herds that they buy from prevent the negative consequences of disease (Suit-B
rather than be in the most self-protective behavior et al., 2020; Doidge et al., 2021). Habit could also be a
classes. Testing and checking herd status are both be- factor and habitually getting vaccinated is important in
haviors associated with BVD accreditation schemes, so human vaccination behavior (Pot et al., 2017). Habit is
these farmers may have bought into the importance of often a barrier to changing farmer behavior (Coyne et
BVD control for the whole farming community. Most al., 2020) and some farmers in the focus groups viewed
of the farmers surveyed invested less than the 50% that the BVD vaccine as insurance. Vaccination could there-
is typical in other research (Johnson and Mislin, 2011); fore be being used both habitually once a farmer starts
Journal of Dairy Science Vol. 105 No. 4, 2022
 Prosser et al.: FARMER PSYCHOSOCIAL PROFILES AND BVD CONTROL 3571

vaccinating and as a protection against the worry of a may be to capitalize on conditional cooperation effects
BVD outbreak. Both motivation factors were correlated (Fischbacher et al., 2001). Conditional cooperation oc-
with the “controlling disease for self and other” factor, curs when people are aware that others are also cooper-
which is unsurprising because motivation is closely as- ating, and this increases the probability that they will
sociated with behavior in the COM-B framework and cooperate (Rustagi et al., 2010). This requires making
farmers who were proactively controlling disease would others’ cooperation behavior observable (Bradley et al.,
have scored highly in the “controlling disease for self 2018) and has been effectively implemented using social
and other” factor (Michie et al., 2011). media (Cameron et al., 2013).
Farmers who felt that they understood how and why In conclusion, psychosocial factors are important for
they should control infectious disease (psychological ca- UK cattle farmer uptake of BVD control. Psychological
pability) and had the time and money to do so (physi- proximity to veterinarians was a novel factor associated
cal opportunity) were more likely to keep a closed herd, with proactive BVD control and was more important
separated from neighboring stock, and undertake BVD here than the more extensively investigated trust. In
testing. Although maintaining a closed, separate herd is addition, lack of trust in other farmers, a high under-
very effective at preventing many infectious diseases, it standing of how and why to control infectious disease,
is not very practical for farmers (Shortall et al., 2017). time and money, and both automatic and reflective
Psychological capability and physical opportunity were motivation were also associated with farmers’ approach
correlated with each other and with automatic and to BVD control. These findings highlight the impor-
reflective motivation and form the context for farmer tance of a close veterinarian-farmer relationship and
behavior; therefore, they all need to be taken into ac- are important for promoting effective BVD control by
count when considering behavior change (Michie et al., farmers, which has implications for successful nation-
2011). Psychological capability had a greater effect on wide BVD control and eradication schemes.
behavior than physical opportunity, suggesting that
knowledge and understanding of how and why to pre-
ACKNOWLEDGMENTS
vent disease is more important for maintaining a closed
and separate herd than time or money, perhaps because All authors acknowledge the Biotechnology and
farmers often appreciate the economic benefits of pre- Biological Sciences Research Council (Swindon, Wilt-
venting and controlling disease in their cattle (Oliveira shire, UK, grant numbers BB/5016341/1 and BB/
et al., 2018; Robinson, 2020). There has been substan- S01750X/1) for funding this project. The funders had
tial research into how to encourage farmers to uptake no other role in the data collection, analysis, or decision
disease control behavior and this finding highlights the to publish. The authors also thank the participating
importance of effective knowledge transfer to farmers, farmers in the focus groups and the survey, the Agricul-
but this needs to be in combination with ensuring that ture and Horticulture Development Board (Kenilworth,
farmers have the physical resources to carry out the Warwickshire, UK), especially Derek Armstrong and
behavior. Jenny Gibbons, for collaborating on and disseminating
Our findings are likely to be generalizable to other the survey, and all the groups and organizations that
similar endemic diseases and farmers in other countries promoted the survey. The authors have not stated any
with similar experiences of veterinarians, government, conflicts of interest.
and neighbors; however, further study will be needed to
investigate this. The farmers in the study were biased
toward English farmers (73% were English compared REFERENCES
with 48% in the national population; Agriculture and
Horticulture Development Board, 2019), therefore, the Abernethy, D. A., J. Moscard-Costello, E. Dickson, R. Harwood, K.
Burns, E. McKillop, S. McDowell, and D. U. Pfeiffer. 2011. Epide-
psychosocial profiles of the general cattle farmer popu- miology and management of a bovine brucellosis cluster in North-
lation may be slightly different. However, country was ern Ireland. Prev. Vet. Med. 98:223–229. https:​/​/​doi​.org/​10​.1016/​
not a confounder in the multinomial model. Respon- j​.prevetmed​.2010​.11​.002.
Agriculture and Horticulture Development Board. 2019. The UK
dents may also have been more interested in infectious Cattle Yearbook 2019. Agriculture and Horticulture Development
diseases than the general farming population; however, Board.
12% of farmers were still doing very little if anything to Aron, A., E. N. Aron, and D. Smollan. 1992. Inclusion of other in the
self scale and the structure of interpersonal closeness. J. Pers. Soc.
control BVD and differences were found in both farmer Psychol. 63:596–612. https:​/​/​doi​.org/​10​.1037/​0022​-3514​.63​.4​.596.
attitudes and BVD control practices. Azbel-Jackson, L., C. Heffernan, G. Gunn, and J. Brownlie. 2018.
From these findings, we recommend that trust and Exploring the role of voluntary disease schemes on UK farmer
bio-security behaviours: Findings from the Norfolk-Suffolk Bovine
proximity to veterinarians and farmers may be crucial Viral Diarrhoea control scheme. PLoS One 13:e0179877. https:​/​/​
to enhancing infection control. One way to achieve this doi​.org/​10​.1371/​journal​.pone​.0179877.

Journal of Dairy Science Vol. 105 No. 4, 2022

 Prosser et al.: FARMER PSYCHOSOCIAL PROFILES AND BVD CONTROL 3572
Bard, A. M., D. Main, E. Roe, A. Haase, H. R. Whay, and K. K. Rey- Ellis-Iversen, J., A. J. C. Cook, E. Watson, M. Nielen, L. Larkin, M.
her. 2019. To change or not to change? Veterinarian and farmer Wooldridge, and H. Hogeveen. 2010. Perceptions, circumstances
perceptions of relational factors influencing the enactment of vet- and motivators that influence implementation of zoonotic control
erinary advice on dairy farms in the United Kingdom. J. Dairy Sci. programs on cattle farms. Prev. Vet. Med. 93:276–285. https:​/​/​doi​
102:10379–10394. https:​/​/​doi​.org/​10​.3168/​jds​.2019​-16364. .org/​10​.1016/​j​.prevetmed​.2009​.11​.005.
Barker, F., L. Atkins, and S. de Lusignan. 2016. Applying the COM-B Engel, C. 2011. Dictator games: A meta study. Exp. Econ. 14:583–610.
behaviour model and behaviour change wheel to develop an inter- https:​/​/​doi​.org/​10​.1007/​s10683​-011​-9283​-7.
vention to improve hearing-aid use in adult auditory rehabilita- Fagerland, M. W., and D. W. Hosmer. 2012. A generalized Hos-
tion. Int. J. Audiol. 55(suppl 3):S90–S98. https:​/​/​doi​.org/​10​.3109/​ mer-Lemeshow goodness-of-fit test for multinomial logistic re-
14992027​.2015​.1120894. gression models. Stata J. 12:447–453. https:​/​/​doi​.org/​10​.1177/​
Bauch, C. T., and D. J. D. Earn. 2004. Vaccination and the theory of 1536867X1201200307.
games. Proc. Natl. Acad. Sci. USA 101:13391–13394. https:​/​/​doi​ Fehr, E., and U. Fischbacher. 2003. The nature of human altruism.
.org/​10​.1073/​pnas​.0403823101. Nature 425:785–791. https:​/​/​doi​.org/​10​.1038/​nature02043.
Berg, J., J. Dickhaut, and K. McCabe. 1995. Trust, reciprocity, and Ferguson, E. 2022. What blood and organ donation can tell us about
social-history. Games Econ. Behav. 10:122–142. https:​/​/​doi​.org/​10​ cooperation? Curr. Opin. Psychol. 44:202–207. https:​/​/​doi​.org/​10​
.1006/​game​.1995​.1027. .1016/​j​.copsyc​.2021​.09​.005.
Bobak, A., and T. Raupach. 2018. Effect of a short smoking cessation Ferguson, E., F. Atsma, W. de Kort, and I. Veldhuizen. 2012. Explor-
training session on smoking cessation behavior and its determi- ing the pattern of blood donor beliefs in first-time, novice, and
nants among general practitioner trainees in England. Nicotine experienced donors: Differentiating reluctant altruism, pure altru-
Tob. Res. 20:1525–1528. https:​/​/​doi​.org/​10​.1093/​ntr/​ntx241. ism, impure altruism, and warm glow. Transfusion 52:343–355.
Böhm, R., C. Betsch, and L. Korn. 2016. Selfish-rational non-vaccina- https:​/​/​doi​.org/​10​.1111/​j​.1537​-2995​.2011​.03279​.x.
tion: Experimental evidence from an interactive vaccination game. Ferguson, E., K. Zhao, and L. D. Smillie. 2020. Personality, prefer-
J. Econ. Behav. Organ. 131:183–195. https:​/​/​doi​.org/​10​.1016/​j​ ences and socioeconomic behavior. Pages 477–493 in Cambridge
.jebo​.2015​.11​.008. Handbook of Personality Psychology, 2nd ed. P. J. Corr and G.
Bradley, A., C. Lawrence, and E. Ferguson. 2018. Does observability Matthews, ed. Cambridge University Press.
affect prosociality? Proc. Biol. Sci. 285:20180116. https:​/​/​doi​.org/​ Fischbacher, U., S. Gachter, and E. Fehr. 2001. Are people condition-
10​.1098/​rspb​.2018​.0116. ally cooperative? Evidence from a public goods experiment. Econ.
Braun, V., and V. Clarke. 2006. Using thematic analysis in psychol- Lett. 71:397–404. https:​/​/​doi​.org/​10​.1016/​S0165​-1765(01)00394​-9.
ogy. Qual. Res. Psychol. 3:77–101. https:​/​/​doi​.org/​10​.1191/​ Fisher, R. 2013. ‘A gentleman’s handshake’: The role of social capital
1478088706qp063oa. and trust in transforming information into usable knowledge. J.
Brennan, M. L., and R. M. Christley. 2013. Cattle producers’ percep- Rural Stud. 31:13–22. https:​/​/​doi​.org/​10​.1016/​j​.jrurstud​.2013​.02​
tions of biosecurity. BMC Vet. Res. 9:71. https:​/​/​doi​.org/​10​.1186/​ .006.
1746​-6148​-9​-71. Forsythe, R., J. L. Horowitz, N. E. Savin, and M. Sefton. 1994. Fair-
Bshary, R., and R. Bergmüller. 2008. Distinguishing four fundamental ness in simple bargaining experiments. Games Econ. Behav. 6:347–
approaches to the evolution of helping. J. Evol. Biol. 21:405–420. 369. https:​/​/​doi​.org/​10​.1006/​game​.1994​.1021.
https:​/​/​doi​.org/​10​.1111/​j​.1420​-9101​.2007​.01482​.x. Fraley, C., and A. E. Raftery. 2002. Model-based clustering, discrimi-
BVDFree. 2021. BVDFree. Accessed Aug. 6, 2021. https:​/​/​bvdfree​.org​ nant analysis, and density estimation. J. Am. Stat. Assoc. 97:611–
.uk/​. 631. https:​/​/​doi​.org/​10​.1198/​016214502760047131.
Cameron, A. M., A. B. Massie, C. E. Alexander, B. Stewart, R. A. Graham, D. A., T. A. Clegg, H. H. Thulke, P. O’Sullivan, G. McGrath,
Montgomery, N. R. Benavides, G. D. Fleming, and D. L. Segev. and S. J. More. 2016. Quantifying the risk of spread of bovine vi-
2013. Social media and organ donor registration: The Facebook ral diarrhoea virus (BVDV) between contiguous herds in Ireland.
effect. Am. J. Transplant. 13:2059–2065. https:​/​/​doi​.org/​10​.1111/​ Prev. Vet. Med. 126:30–38. https:​/​/​doi​.org/​10​.1016/​j​.prevetmed​
ajt​.12312. .2016​.01​.017.
Carroll, G. A., and J. M. Groarke. 2019. The importance of the social Gunn, G. J., H. W. Saatkamp, R. W. Humphry, and A. W. Stott.
sciences in reducing tail biting prevalence in pigs. Animals (Basel) 2005. Assessing economic and social pressure for the control of
9:591. https:​/​/​doi​.org/​10​.3390/​ani9090591. bovine viral diarrhoea virus. Prev. Vet. Med. 72:149–162. https:​/​/​
Charness, G., U. Gneezy, and B. Halladay. 2016. Experimental meth- doi​.org/​10​.1016/​j​.prevetmed​.2005​.08​.012.
ods: Pay one or pay all. J. Econ. Behav. Organ. 131:141–150. https:​ Gwaredu BVD. 2019. Animal Health & Welfare Wales. Accessed Aug.
/​/​doi​.org/​10​.1016/​j​.jebo​.2016​.08​.010. 6, 2021. https:​/​/​www​.ahww​.cymru/​en/​.
Chetty, R., A. Hofmeyr, H. Kincaid, and B. Monroe. 2021. The trust Hardefeldt, L. Y., J. R. Gilkerson, H. Billman-Jacobe, M. A. Steven-
game does not (only) measure trust: The risk-trust confound re- son, K. Thursky, K. E. Bailey, and G. F. Browning. 2018. Barriers
visited. J. Behav. Exp. Econ. 90:101520. https:​/​/​doi​.org/​10​.1016/​ to and enablers of implementing antimicrobial stewardship pro-
j​.socec​.2020​.101520. grams in veterinary practices. J. Vet. Intern. Med. 32:1092–1099.
Cialdini, R. B., S. L. Brown, B. P. Lewis, C. Luce, and S. L. Neuberg. https:​/​/​doi​.org/​10​.1111/​jvim​.15083.
1997. Reinterpreting the empathy-altruism relationship: When one Heffernan, C., L. Azbel-Jackson, J. Brownlie, and G. Gunn. 2016.
into one equals oneness. J. Pers. Soc. Psychol. 73:481–494. https:​/​ Farmer attitudes and livestock disease: exploring citizenship be-
/​doi​.org/​10​.1037/​0022​-3514​.73​.3​.481. haviour and peer monitoring across two BVD control schemes in
Coyne, L. A., C. Bellet, S. M. Latham, and D. Williams. 2020. Provid- the UK. PLoS One 11:e0152295. https:​/​/​doi​.org/​10​.1371/​journal​
ing information about triclabendazole resistance status influences .pone​.0152295.
farmers to change liver fluke control practices. Vet. Rec. 187:357. Heffernan, C., L. Nielsen, K. Thomson, and G. Gunn. 2008. An explo-
https:​/​/​doi​.org/​10​.1136/​vr​.105890. ration of the drivers to bio-security collective action among a sam-
Dimick, M., D. Rueda, and D. Stegmueller. 2018. Models of other- ple of UK cattle and sheep farmers. Prev. Vet. Med. 87:358–372.
regarding preferences, inequality, and redistribution. Annu. Rev. https:​/​/​doi​.org/​10​.1016/​j​.prevetmed​.2008​.05​.007.
Polit. Sci. 21:441–460. https:​/​/​doi​.org/​10​.1146/​annurev​-polisci​ Hernández-Jover, M., V. Higgins, M. Bryant, L. Rast, and C. Mc-
-091515​-030034. Shane. 2016. Biosecurity and the management of emergency ani-
Doidge, C., E. Ferguson, F. Lovatt, and J. Kaler. 2021. Understanding mal disease among commercial beef producers in New South Wales
farmers’ naturalistic decision making around prophylactic antibi- and Queensland (Australia). Prev. Vet. Med. 134:92–102. https:​/​/​
otic use in lambs using a grounded theory and natural language doi​.org/​10​.1016/​j​.prevetmed​.2016​.10​.005.
processing approach. Prev. Vet. Med. 186:105226. https:​/​/​doi​.org/​ Houe, H. 1999. Epidemiological features and economical importance
10​.1016/​j​.prevetmed​.2020​.105226. of bovine virus diarrhoea virus (BVDV) infections. Vet. Microbiol.
64:89–107. https:​/​/​doi​.org/​10​.1016/​S0378​-1135(98)00262​-4.

Journal of Dairy Science Vol. 105 No. 4, 2022

 Prosser et al.: FARMER PSYCHOSOCIAL PROFILES AND BVD CONTROL 3573
Hu, L. T., and P. M. Bentler. 1999. Cutoff criteria for fit indexes in R Core Team. 2019. R: A language and environment for statistical
covariance structure analysis: Conventional criteria versus new al- computing. R Foundation for Statistical Computing.
ternatives. Struct. Equ. Modeling 6:1–55. https:​/​/​doi​.org/​10​.1080/​ Rand, D. G., and M. A. Nowak. 2013. Human cooperation. Trends
10705519909540118. Cogn. Sci. 17:413–425. https:​/​/​doi​.org/​10​.1016/​j​.tics​.2013​.06​.003.
Johnson, N. D., and A. A. Mislin. 2011. Trust games: A meta-analysis. Revelle, W. 2020. psych: Procedures for Personality and Psychological
J. Econ. Psychol. 32:865–889. https:​/​/​doi​.org/​10​.1016/​j​.joep​.2011​ Research. 2.0.8 ed. Northwestern University.
.05​.007. Robinson, P. A. 2020. “They’ve got to be testing and doing some-
Kleinert, T., B. Schiller, U. Fischbacher, L. A. Grigutsch, N. Koranyi, thing about it”: Farmer and veterinarian views on drivers for
K. Rothermund, and M. Heinrichs. 2020. The Trust Game for Johne’s disease control in dairy herds in England. Prev. Vet. Med.
Couples (TGC): A new standardized paradigm to assess trust in 182:105094. https:​/​/​doi​.org/​10​.1016/​j​.prevetmed​.2020​.105094.
romantic relationships. PLoS One 15:e0230776. https:​/​/​doi​.org/​10​ Rustagi, D., S. Engel, and M. Kosfeld. 2010. Conditional cooperation
.1371/​journal​.pone​.0230776. and costly monitoring explain success in forest commons man-
Kong, D. T. 2018. Trust toward a group of strangers as a function of agement. Science 330:961–965. https:​/​/​doi​.org/​10​.1126/​science​
stereotype-based social identification. Pers. Individ. Dif. 120:265– .1193649.
270. https:​/​/​doi​.org/​10​.1016/​j​.paid​.2017​.03​.031. Scrucca, L., M. Fop, T. B. Murphy, and A. E. Raftery. 2016. mclust
Lee, D. 2013. CARBayes: an R package for Bayesian spatial model- 5: Clustering, classification and density estimation using Gaussian
ing with conditional autoregressive priors. J. Stat. Softw. 55:1–24. finite mixture models. R J. 8:289–317. https:​/​/​doi​.org/​10​.32614/​
https:​/​/​doi​.org/​10​.18637/​jss​.v055​.i13. RJ​-2016​-021.
Mashek, D., L. W. Cannaday, and J. P. Tangney. 2007. Inclusion of Shortall, O., M. Green, M. Brennan, W. Wapenaar, and J. Kaler.
community in self scale: A single-item pictorial measure of com- 2017. Exploring expert opinion on the practicality and effective-
munity connectedness. J. Community Psychol. 35:257–275. https:​/​ ness of biosecurity measures on dairy farms in the United King-
/​doi​.org/​10​.1002/​jcop​.20146. dom using choice modeling. J. Dairy Sci. 100:2225–2239. https:​/​/​
McClurkin, A. W., E. T. Littledike, R. C. Cutlip, G. H. Frank, M. F. doi​.org/​10​.3168/​jds​.2016​-11435.
Coria, and S. R. Bolin. 1984. Production of cattle immunotolerant Shortall, O., A. Ruston, M. Green, M. Brennan, W. Wapenaar, and J.
to bovine viral diarrhea virus. Can. J. Comp. Med. 48:156–161. Kaler. 2016. Broken biosecurity? Veterinarians’ framing of biosecu-
McKnight, D. H., and N. L. Chervany. 2001. Trust and Distrust Defi- rity on dairy farms in England. Prev. Vet. Med. 132:20–31. https:​
nitions: One bite at a time. Pages 27–54 in Trust in Cyber-Societ- /​/​doi​.org/​10​.1016/​j​.prevetmed​.2016​.06​.001.
ies: Integrating the Human and Artificial Perspectives. R. Falcone, Suit-B, Y., L. Hassan, S. E. Krauss, S. Z. Ramanoon, P. T. Ooi,
M. Singh, and Y.-H. Tan, ed. Springer. A. Yasmin, and J. Epstein. 2020. Exploring the mental model of
Michie, S., L. Atkins, and R. West. 2014. The Behaviour Change cattle farmers in disease prevention and control practices. Vet. Sci.
Wheel: A guide to designing interventions. Silverback Publishing. 7:27. https:​/​/​doi​.org/​10​.3390/​vetsci7010027.
Michie, S., M. M. van Stralen, and R. West. 2011. The behaviour Taylor, M. J., D. Arriscado, I. Vlaev, D. Taylor, P. Gately, and A. Dar-
change wheel: A new method for characterising and designing be- zi. 2016. Measuring perceived exercise capability and investigating
haviour change interventions. Implement. Sci. 6:42. https:​/​/​doi​ its relationship with childhood obesity: A feasibility study. Int.
.org/​10​.1186/​1748​-5908​-6​-42. J. Obes. (Lond.) 40:34–38. https:​/​/​doi​.org/​10​.1038/​ijo​.2015​.210.
Murphy, R. O., and K. A. Ackermann. 2014. Social value orientation: Thurstone, L. L. 1947. Multiple-factor analysis: A development and
Theoretical and measurement issues in the study of social prefer- expansion of the vectors of the mind. The University of Chicago
ences. Pers. Soc. Psychol. Rev. 18:13–41. https:​/​/​doi​.org/​10​.1177/​ Press.
1088868313501745. Tu, K. C., S. Chen, and R. M. Mesler. 2021. Trait self-construal, in-
Murphy, R. O., K. A. Ackermann, and M. J. J. Handgraaf. 2011. clusion of others in the self and self-control predict stay-at-home
Measuring social value orientation. Judgm. Decis. Mak. 6:771–781. adherence during COVID-19. Pers. Individ. Dif. 175:110687. https:​
Nowak, M. A. 2006. Five rules for the evolution of cooperation. Science /​/​doi​.org/​10​.1016/​j​.paid​.2021​.110687.
314:1560–1563. https:​/​/​doi​.org/​10​.1126/​science​.1133755. Venables, W. N., and B. D. Ripley. 2002. Modern Applied Statistics
O’Kane, H., E. Ferguson, J. Kaler, and L. Green. 2017. Associations with S. 4th ed. Springer.
between sheep farmer attitudes, beliefs, emotions and personal- Vernon, M. C. 2011. Demographics of cattle movements in the United
ity, and their barriers to uptake of best practice: The example Kingdom. BMC Vet. Res. 7:31. https:​/​/​doi​.org/​10​.1186/​1746​-6148​
of footrot. Prev. Vet. Med. 139:123–133. https:​/​/​doi​.org/​10​.1016/​ -7​-31.
j​.prevetmed​.2016​.05​.009. Weldegebriel, H. T., G. J. Gunn, and A. W. Stott. 2009. Evaluation
Oliveira, V. H. S., I. Anneberg, H. Voss, J. T. Sorensen, and P. T. of producer and consumer benefits resulting from eradication of
Thomsen. 2018. Attitudes of Danish dairy farmers towards bios- bovine viral diarrhoea (BVD) in Scotland, United Kingdom. Prev.
ecurity. Livest. Sci. 214:153–160. https:​/​/​doi​.org/​10​.1016/​j​.livsci​ Vet. Med. 88:49–56. https:​/​/​doi​.org/​10​.1016/​j​.prevetmed​.2008​.07​
.2018​.06​.004. .001.
Peterhans, E., T. W. Jungi, and M. Schweizer. 2003. BVDV and in- West, S. A., A. S. Griffin, and A. Gardner. 2007. Social semantics:
nate immunity. Biologicals 31:107–112. https:​/​/​doi​.org/​10​.1016/​ Altruism, cooperation, mutualism, strong reciprocity and group
S1045​-1056(03)00024​-1. selection. J. Evol. Biol. 20:415–432. https:​/​/​doi​.org/​10​.1111/​j​.1420​
Pfattheicher, S., Y. A. Nielsen, and I. Thielmann. 2022. Prosocial be- -9101​.2006​.01258​.x.
havior and altruism: A review of concepts and definitions. Curr.
Opin. Psychol. 44:124–129. https:​/​/​doi​.org/​10​.1016/​j​.copsyc​.2021​
.08​.021. ORCIDS
Pletzer, J. L., D. Balliet, J. Joireman, D. M. Kuhlman, S. C. Voelpel,
and P. A. M. Van Lange. 2018. Social value orientation, expecta- N. S. Prosser https:​/​/​orcid​.org/​0000​-0001​-8391​-7153
tions, and cooperation in social dilemmas: A meta-analysis. Eur. M. J. Green https:​/​/​orcid​.org/​0000​-0002​-6408​-6443
J. Pers. 32:62–83. https:​/​/​doi​.org/​10​.1002/​per​.2139. E. Ferguson https:​/​/​orcid​.org/​0000​-0002​-7678​-1451
Pot, M., H. M. van Keulen, R. A. C. Ruiter, I. Eekhout, L. Mollema, M. J. Tildesley https:​/​/​orcid​.org/​0000​-0002​-6875​-7232
and T. Paulussen. 2017. Motivational and contextual determinants
of HPV-vaccination uptake: A longitudinal study among mothers E. M. Hill https:​/​/​orcid​.org/​0000​-0002​-2992​-2004
of girls invited for the HPV-vaccination. Prev. Med. 100:41–49. M. J. Keeling https:​/​/​orcid​.org/​0000​-0003​-4639​-4765
https:​/​/​doi​.org/​10​.1016/​j​.ypmed​.2017​.04​.005. J. Kaler https:​/​/​orcid​.org/​0000​-0002​-3332​-7064

Journal of Dairy Science Vol. 105 No. 4, 2022