If you have made it to Phase 2, you should now have a broad topic in mind that needs to be shaped into a feasible research study. This phase focuses on refining that topic into a clearly articulated research question and then developing a strong supporting rationale. Together, these elements should explain not only what you want to study, but why the answer matters and how it has the potential to improve outcomes in veterinary medicine, animal welfare or production systems, or society more broadly.
More than 30 different frameworks have been developed to help researchers work through the process of writing good research questions. Most share common elements of getting you to think about who you are studying, what you are studying, and how you are measuring or describing the outcomes. This section will cover five of the most popular frameworks for most situations in animal health and veterinary research.
Quick guide to choosing a research question framework
The Four Questions framework is best used early in project development or when a study does not fit neatly into an intervention–comparison structure. It is intentionally flexible and can be applied across quantitative, qualitative, descriptive, and mixed-methods research.
Rather than prescribing a specific study design, this framework helps clarify what you are studying, who it applies to, and what you hope to learn. This framework is often used as a starting point, with the resulting question later refined using more structured frameworks such as PICO or SPIDER once the study focus becomes clearer.
This defines the phenomenon you want to study and helps narrow a broad topic into a specific, researchable problem. At this stage, the focus is on clearly identifying the topic or issue that motivates the research.
This identifies the population, system, or setting of interest. This may include a particular species, production system, clinical population, geographic region, or stakeholder group. This question provides the context for the research.
This clarifies the nature of the inquiry. You may be evaluating an intervention, comparing groups, examining an exposure, or exploring experiences or processes. Not all studies require a formal comparator, but this question forces you to be explicit about what you are doing. Common verbs at this stage include to explore, to describe, to explain, to identify, or to understand.
This specifies what you want to measure, estimate, describe, or understand, and links the research question to its intended impact. Outcomes may relate to practice, policy, animal welfare, production efficiency, or broader knowledge generation.
Example uses:
Example question:
What factors contribute to delayed veterinary presentation for lameness in pasture-based dairy herds, and how do farm management practices influence this, to inform earlier intervention strategies?
The PICO framework is best used when you are evaluating the effect of an intervention and want a clearly testable, quantitative research question. It is particularly well suited to clinical trials, intervention studies, epidemiological analyses, and evidence-based practice questions where comparison and measurement are central.
PICO helps ensure that the key components of a focused research question are explicit and logically connected.
P: Population or problem
The specific population, disease, or condition of interest. This may include characteristics such as species, age group, production system, clinical diagnosis, or other defining features.
I: Intervention or exposure
The intervention, treatment, management strategy, or exposure being investigated. This may also include a risk factor or prognostic variable, depending on the study type.
C: Comparator
The alternative against which the intervention or exposure is compared. This may be another treatment, placebo, standard practice, absence of exposure, or a different risk or prognostic factor.
O: Outcome
The outcome of interest, such as disease incidence, treatment response, diagnostic accuracy, production measure, welfare indicator, or adverse event.
T: Time
The timeframe over which the outcome is measured or observed. This may be explicitly stated or implied elsewhere in the question or study design.
Example uses:
Example question:
In adult dairy cows with clinical mastitis, does intramammary antibiotic A compared with antibiotic B reduce time to clinical cure within 7 days?
Common PICO question templates by study purpose
Used to examine whether an exposure or characteristic is associated with increased or decreased risk of an outcome.
Are ______ (Population) who have ______ (Exposure) at ______ (increased/decreased) risk of ______ (Outcome) compared with ______ (Population) with or without ______ (Comparator) over ______ (Time)?
Used when evaluating the effect of a treatment, management strategy, or intervention.
In ______ (Population), what is the effect of ______ (Intervention) on ______ (Outcome) compared with ______ (Comparator) within ______ (Time)?
Used to evaluate the accuracy or usefulness of a diagnostic test or method.
Is ______ (Index test) more accurate in diagnosing ______ (Population or condition) compared with ______ (Comparator test) for ______ (Outcome)?
Prognosis or prediction
Used to examine how a factor influences future outcomes in individuals with a defined condition.
Does ______ (Prognostic factor) influence ______ (Outcome) in patients who have ______ (Population or condition) over ______ (Time)?
Used for qualitative research exploring perceptions, experiences, or meanings. While PICO is less commonly used for qualitative studies, the population and phenomenon of interest can still be clearly specified.
How do ______ (Population) diagnosed with ______ (Condition or experience) perceive ______ (Phenomenon of interest) during ______ (Time or context)?
Note: Not every research question will include all PICO elements. For example, some questions do not involve a true intervention, and in some cases the time component is implicit rather than explicitly stated.
The SPIDER framework is best used when the research focus is on experiences, perceptions, behaviours, or decision-making rather than on measuring the effect of a defined intervention. It is particularly well suited to qualitative and mixed-methods research, including interviews, focus groups, surveys, and participatory studies commonly used to explore behaviour change, engagement, and implementation challenges in the dairy sector.
SPIDER helps ensure that the population of interest, the phenomenon being explored, and the type of evidence being sought are clearly articulated, supporting the development of research questions that are transparent, coherent, and well aligned with the aims of qualitative inquiry.
S: Sample
Who is being studied. This refers to the specific group of people or stakeholders of interest, such as animal owners, farmers, veterinarians, or policy actors.
PI: Phenomenon of Interest
What experience, behaviour, decision, or process is being explored. The focus is on what participants think, feel, or do, rather than on intervention effects.
D: Design
How the data are collected. Common approaches include interviews, focus groups, observations, or open-ended surveys.
E: Evaluation
What is being explored or understood as the outcome of interest, such as experiences, attitudes, perceptions, motivations, or decision-making processes.
R: Research type
The methodological approach, typically qualitative or mixed methods.
Example uses:
Example question:
How do cat owners perceive the risks and benefits of containment, explored through focus groups, in terms of attitudes and decision-making, using qualitative methods?
The SPICE framework is best used for applied, real-world research where setting, perspective, and impact are central. It is particularly suited to service delivery, education, programme evaluation, and practice-based research.
SPICE focuses attention on context and practical outcomes, making it useful for studies intended to inform implementation or decision-making in real-world settings.
S: Setting
The context in which the issue occurs, such as a geographic location, service environment, or organisational context.
P: Perspective
The users, potential users, or stakeholders of the service or intervention, such as clients, professionals, or communities.
I: Intervention / Interest / Exposure
The action, service, programme, or change being examined.
C: Comparison
The alternative action, service, or status quo against which the intervention is compared.
E: Evaluation
The outcome or measure used to assess impact, effectiveness, or success.
Example uses:
Example question:
In rural veterinary practices, how do production-animal veterinarians experience the introduction of tele-advisory services compared with traditional in-person consultations, in terms of workflow efficiency and client satisfaction?
The ECLIPSE framework is best used for health services, organisational, or policy research rather than clinical effectiveness studies. It is particularly helpful when the “intervention” is a service, system, or policy rather than a treatment delivered to individual patients.
ECLIPSE is designed to support questions focused on service improvement, system design, and policy development.
E: Expectation
What you are aiming to improve, change, or inform, and how the information will be used. This clarifies the purpose of the research and its intended application.
C: Client group
Who the service or policy is intended for, such as patients, clients, animal owners, or communities.
L: Location
Where the service or policy operates, for example within hospitals, veterinary practices, regional services, or national systems.
I: Impact
The change or outcome the research is investigating, such as improved access, efficiency, quality, safety, or wellbeing.
P: Professionals
Who is involved in delivering, managing, or improving the service or policy, such as clinicians, administrators, regulators, or support staff.
S: Service
The type of service, system, or policy being examined.
Example uses:
Example question:
How could after-hours emergency services for companion animals in regional New Zealand be redesigned to reduce clinician burnout and improve continuity of care for veterinarians and nurses?
Ambiguous research questions are difficult to answer, difficult to interpret, and often lead to poor-quality studies. Ambiguity usually arises from vague terms, poorly defined populations, unclear timing, or imprecise outcomes. At this stage, your goal is to make every key component of the question explicit.
Initial question:
Do thin cows produce less milk?
Problems:
The term “thin” is subjective and undefined. It is unclear how thinness is measured and at what point in time. The term “cow” is also ambiguous and could include different classes of cattle, such as dairy cows or heifers.
Improved question:
Do cows with a body condition score of less than or equal to 3.5 at the time of calving produce less milk?
This version removes ambiguity by defining body condition using a recognised scoring system, specifying a threshold, and clearly stating when body condition is assessed.
Initial question:
Does feeding a high-protein diet improve growth in calves?
Problems:
“High-protein” is undefined and could refer to a wide range of diets. “Growth” is also vague and may be measured in several different ways or over different timeframes. The age range and production system of the calves are not specified.
Improved question:
When compared with a diet containing 16% crude protein, does feeding a diet containing 20% crude protein increase average daily liveweight gain in dairy calves between 4 and 12 weeks of age?
This version removes ambiguity by clearly defining the diets being compared, specifying how growth is measured, and stating the age range of interest.
As a general rule, ask yourself whether an independent researcher could read your question and design the same study you are imagining. If not, further clarification is needed.
Once you have a clearly articulated research question, the next step is to develop a strong research rationale. This forms the backbone of the introduction section of a scientific manuscript and explains why the study is needed and why it is worth doing. A well-written rationale moves from a broad description of the problem to a specific justification for your study, clearly linking the research question to meaningful outcomes.
Begin by clearly describing the problem, challenge, or opportunity that motivates the research. This may relate to suboptimal clinical outcomes, animal welfare concerns, inefficiencies in production systems, gaps in service delivery, or emerging risks or opportunities for improvement. The problem should be grounded in real-world practice or policy, not just academic curiosity.
Ask yourself: What is currently not working as well as it could, or what opportunity exists to improve outcomes?
Next, explain who is impacted by this problem and in what ways. This may include animals, animal owners, veterinarians, producers, industry bodies, regulators, or society more broadly. Be explicit about the nature of the impact, such as welfare compromise, economic loss, health risks, ethical concerns, workload pressures, or inequities in access to services.
This step helps demonstrate why the problem matters and for whom the research has value.
Then summarise what is already being done to address the problem and why these approaches are not fully adequate. This may involve limitations in current guidelines, inconsistent evidence, poor uptake in practice, contextual constraints, or unintended consequences of existing interventions.
The goal here is not to dismiss prior work, but to show that despite existing efforts, important limitations remain.
From this context, clearly identify the specific gap in knowledge that your study will address. This is the point where the rationale narrows from the general problem to the precise issue your research is designed to resolve. The gap may relate to a lack of data in a particular population, uncertainty about effectiveness, insufficient understanding of mechanisms, or limited evidence under real-world conditions.
A strong knowledge gap statement makes it clear why your study is necessary now.
Next, explain how the findings will be used and by whom. This might include informing clinical decision-making, improving management guidelines, supporting policy development, refining welfare assessments, or guiding future research. Being explicit about end users helps ensure the research is designed with impact in mind.
This step also helps align your study with relevance and translational value.
Finally, link the rationale directly back to the research question by briefly clarifying the key elements of the study. This includes the population of interest, the intervention or exposure being examined, the comparator (if relevant), and the outcomes that matter most. These elements should clearly follow from the problem and knowledge gap you have described.
By the end of the research rationale, the reader should understand exactly why the study is being conducted, who it is for, what gap it fills, and how answering the research question will contribute to improved outcomes.
Before moving on to study design and methods, it is important to pause and critically review your research question and rationale. The FINER framework provides a high-level check of whether the research question is worth pursuing, evaluating the overall quality and suitability as a research project.
Can the study realistically be done with the time, resources, data, expertise, and access available? Consider sample size, data availability, costs, and logistical constraints.
Is the question interesting to you and to others in the field? A question that sustains motivation is important, particularly for longer projects.
Does the study add something new? This may be entirely new knowledge, a new population, a new context, or a better-designed test of an existing idea.
Can the study be conducted ethically? This includes animal welfare considerations, informed consent where relevant, and compliance with ethical review requirements.
Will answering the question improve veterinary practice, animal welfare, production efficiency, policy, or societal outcomes? Relevance is a key justification for undertaking research.
If a question struggles to meet multiple FINER criteria, it may need to be refined, narrowed, or reconsidered before progressing further. I also find it useful to consider how you imagine people using the results from your research in the real world to improve current outcomes. Is it re-writing policy, making different decisions about treating clinical cases, or simply having improved awareness of a topic? Knowing what changes you want people to make can help guide your choices for study design to ensure the results are meaningful and pratical for those you are trying to reach.
Advancing Animal Care through Shared Learning