# Conducting Systematic Reviews of Veterinary Therapeutic Interventions


## Key Takeaways

- Veterinary therapeutic systematic reviews necessitate a PICO framework (Population, Intervention, Comparator, Outcome) adapted for species, breed, and study design, with protocol registration (e.g., CAMARADES for animal studies) recommended prior to screening to mitigate bias.
- Comprehensive literature searches should include core biomedical databases (MEDLINE/PubMed, Embase, Web of Science) augmented by species-specific (e.g., CAB Abstracts) and grey literature sources to capture the full evidence base.
- Risk of bias assessment is critical, employing tools like SYRCLE's for animal intervention studies and Cochrane RoB 2 for randomized clinical trials, with independent reviewers judging domains such as blinding and incomplete outcome data.
- Data synthesis requires careful consideration of heterogeneity; meta-analysis is appropriate only with sufficient clinical and methodological homogeneity, otherwise, qualitative synthesis is mandated, often using structured tables to summarize findings.
- Reporting must adhere to PRISMA 2020 for the review itself, with included animal studies assessed against ARRIVE 2.0 guidelines, ensuring transparency in methods, risk of bias, and data extraction.
- Common failure modes include the mismatch between review questions and heterogeneous animal model data, publication bias due to underrepresentation of negative results, and conflation of disparate study designs (e.g., in vitro vs. in vivo) without stratification.

---

Systematic reviews of therapeutic interventions answer a focused clinical question by locating, appraising, and synthesising all available primary studies that meet predefined eligibility criteria. In veterinary medicine, these reviews serve two distinct audiences. Clinicians use them to judge whether a treatment is supported by sufficient evidence to justify its use in practice. Researchers use them to identify knowledge gaps, estimate effect sizes for sample size calculations, and determine whether further primary studies are ethically justifiable. This article provides a procedural framework for conducting such reviews across species, with emphasis on the specific challenges that arise when the primary literature consists of animal model studies, client-owned patient trials, or a mixture of both.

The need for rigorous evidence synthesis in veterinary therapeutics has grown as the volume of published intervention studies has expanded. A systematic review differs from a narrative review in that its methods are specified in advance, its literature search is reproducible, and its conclusions are tied directly to the quality and quantity of the underlying evidence. The same methodological logic that governs human clinical systematic reviews applies to veterinary reviews, but several features of the veterinary evidence base require adaptation. These include heterogeneous study populations across species and breeds, variable reporting of animal husbandry and housing conditions, and the frequent absence of registered protocols or standardized outcome definitions.

This article covers the full procedural arc of a veterinary therapeutic systematic review: formulating the question, registering the protocol, designing the search, screening and selecting studies, assessing risk of bias, extracting data, synthesising results, and reporting. It addresses both quantitative synthesis through meta-analysis and qualitative synthesis where statistical pooling is inappropriate. Diagnostic test accuracy reviews are excluded, the focus is exclusively on interventions intended to prevent, treat, or manage disease.

## At a Glance

| Parameter | Decision or Standard |
|---|---|
| Review question format | PICO (Population, Intervention, Comparator, Outcome), adapted for species and study design |
| Protocol registration | Recommended before screening begins, animal studies may be registered in CAMARADES |
| Primary databases | MEDLINE/PubMed, Embase, Web of Science, add species-specific and regional databases |
| Risk of bias tool | SYRCLE's tool for animal intervention studies, Cochrane RoB 2 for randomised clinical trials |
| Reporting guideline | PRISMA 2020 for the review itself, ARRIVE 2.0 for assessing reporting quality of included animal studies |
| Meta-analysis precondition | Sufficient clinical and methodological homogeneity, otherwise use narrative synthesis |
| Small-study effects | Assess with funnel plots and Egger's test when at least 10 studies are pooled |
| Reporting standards library | EQUATOR Network for identifying the applicable guideline |

## The Logic of Evidence Synthesis for Therapeutic Questions

A systematic review is an observational study of studies. Its validity depends on the same principles that govern primary research: a prespecified question, a sampling frame that is as complete as possible, and measurement procedures that are applied consistently across all included items. For therapeutic interventions, the review seeks to estimate the direction and magnitude of treatment effect, to assess the precision of that estimate, and to determine whether the effect is consistent across different populations, intervention protocols, and outcome measures.

The hierarchy of evidence places systematic reviews above individual studies because a review can detect patterns that single studies cannot. Small studies with modest effect sizes may individually fail to reach statistical significance, yet collectively demonstrate a consistent treatment benefit. Conversely, a body of small positive studies may reflect publication bias instead of true efficacy. The review's task is to distinguish these possibilities using explicit statistical and qualitative methods.

Veterinary therapeutic reviews face a particular tension. The most directly applicable evidence comes from clinical trials in client-owned animals, but such trials are relatively scarce for many conditions. Animal model studies, particularly rodent models of human disease, are far more numerous and are frequently used to inform the design of subsequent clinical trials. A review restricted to client-owned animal trials may have too few studies for meaningful synthesis. A review that includes animal model studies must confront substantial between-study heterogeneity in species, strain, induction method, timing of intervention, and outcome assessment. The review team must decide in advance which study designs answer the clinical question and must justify that decision in the protocol.

## Formulating the Review Question

The review question determines every subsequent step. A poorly formulated question produces a search that is either too broad, yielding thousands of irrelevant records, or too narrow, missing relevant studies. The PICO framework provides the standard structure. For veterinary therapeutic reviews, each PICO element requires careful specification.

The population element must specify species, and where relevant, breed, age, sex, and reproductive status. It must also specify the disease or condition, including diagnostic criteria and severity. A review of analgesics for canine osteoarthritis, for example, must decide whether to include only dogs with radiographic confirmation of disease or also those diagnosed on clinical signs alone. The intervention element must specify the treatment, its route, dose, and duration, and whether co-interventions are permitted. The comparator element may be placebo, no treatment, an active control, or a different dose of the same intervention. The outcome element must specify which outcomes are of interest, how they are measured, and at what time points.

The review question also requires a decision about study design. Some therapeutic questions can only be answered by randomised controlled trials. Others, particularly those concerning rare adverse effects or long-term outcomes, may require the inclusion of cohort studies or case series. The review team should specify the minimum acceptable design in the protocol and should be prepared to justify exclusions. The [EQUATOR Network reporting guideline library](https://www.equator-network.org/) provides a structured approach to identifying the appropriate reporting standards for each included study design, which in turn informs the eligibility criteria.

## Protocol Development and Registration

A protocol is a written specification of the review methods, completed before the literature search begins. It serves three functions. It reduces the risk of bias arising from post hoc decisions about study selection and analysis. It provides a public record against which the final review can be audited. And it allows other researchers to identify ongoing reviews and avoid duplication.

The protocol should state the review question, the search strategy including all databases and search terms, the eligibility criteria, the screening and selection process, the risk of bias assessment method, the data extraction items, and the planned synthesis approach. It should also specify how heterogeneity will be assessed and what thresholds will trigger a decision to abandon meta-analysis in favour of narrative synthesis.

For reviews that include animal model studies, registration in the Collaborative Approach to Meta-Analysis and Review of Animal Data from Experimental Studies (CAMARADES) database is an established practice. The CAMARADES database was designed specifically to support systematic reviews and meta-analyzes of animal studies and provides a registry that is searchable by other review teams. Reviews of clinical trials in client-owned animals may be registered in the same registries used for human systematic reviews, although some registries have restrictions on non-human studies. The review team should check the eligibility requirements of each registry before committing to a registration venue.

The protocol should also address the question of language restrictions. Many veterinary systematic reviews restrict inclusion to English-language publications, but this can introduce bias, particularly for conditions where substantial research is published in other languages. The protocol should state the language policy and its rationale.

## Search Strategy Development and Execution

The search strategy translates the structured question into database-specific syntax. For veterinary therapeutic reviews, the search must capture both the intervention and the population while remaining sensitive enough to detect studies indexed under inconsistent terminology. Species names, breed terms, and production-system descriptors should be combined with intervention synonyms and outcome-related terms using Boolean operators.

Search at least three bibliographic databases. [PubMed](https://pubmed.ncbi.nlm.nih.gov/25881229/), Embase, and Web of Science form a common core for veterinary questions, as demonstrated in systematic reviews of mesenchymal stem cell therapy for traumatic brain injury and of dorsal root ganglion stimulation for pain. CAB Abstracts and Agricola add veterinary-specific and production-animal coverage that biomedical databases may miss. Scopus and the Cochrane Library may supplement these depending on the question. Grey literature sources include conference proceedings, thesis repositories, and regulatory agency websites.

The search structure should follow the PICO elements. Population terms use controlled vocabulary where available, such as MeSH in PubMed, combined with free-text terms for synonyms and variant spellings. Intervention terms require particular care because therapeutic interventions in veterinary medicine often have multiple names, including trade names, generic names, and procedural descriptors. Outcome terms are frequently omitted from the search to avoid missing studies that report unexpected or secondary outcomes.

Validation of the search strategy is essential. Test the strategy against a set of five to ten known relevant articles identified during protocol development. The strategy must retrieve all of them. If it does not, revise the search terms and repeat the validation. Record the final strategy for each database in full, including the date of execution, so that the review can be updated or replicated.

## Study Selection and Screening

Study selection proceeds in two stages. The first stage screens titles and abstracts against the eligibility criteria. The second stage reviews full texts of potentially eligible records. Both stages require two independent reviewers working in duplicate. Disagreements are resolved by discussion or by a third reviewer acting as arbiter.

Screening decisions should be guided by a piloted eligibility form derived from the protocol's inclusion and exclusion criteria. The form should be brief and operational. For example, a criterion stating "studies must report a comparator group" becomes a screening question: "Does the study include a control or comparator arm?" Reviewers mark each record as include, exclude, or uncertain at each stage.

The screening process must be documented in a PRISMA flow diagram adapted for veterinary studies. The diagram records the number of records identified through database searching, records after duplicate removal, records screened, records excluded at title and abstract stage, full-text articles assessed for eligibility, full-text articles excluded with reasons, and studies included in the qualitative synthesis and meta-analysis. This flow diagram serves as the primary transparency record for the selection process.

## Risk of Bias Assessment

Risk of bias assessment evaluates the internal validity of included studies. For veterinary therapeutic studies, the [Systematic Review Center for Laboratory Animal Experimentation (SYRCLE) risk of bias tool](https://pubmed.ncbi.nlm.nih.gov/29152818/) is the most widely used instrument. It adapts the Cochrane risk of bias domains to animal studies, covering sequence generation, baseline characteriztics, allocation concealment, random housing, blinding of caregivers and investigators, random outcome assessment, blinding of outcome assessment, incomplete outcome data, selective outcome reporting, and other sources of bias.

For clinical veterinary trials, the Cochrane Risk of Bias tool for randomized trials provides an appropriate framework. The tool assesses bias arising from the randomization process, deviations from intended interventions, missing outcome data, measurement of the outcome, and selection of the reported result. Veterinary-specific considerations include the adequacy of blinding when the same clinician administers treatment and assesses outcomes, and the handling of animals withdrawn from trials for humane reasons.

Assessment requires two independent reviewers. Each domain is judged as low risk, high risk, or unclear risk, with the judgment supported by a brief justification drawn from the study text. Disagreements are resolved through discussion. The results should be presented in a table or figure that displays the risk of bias judgments for each study across all domains.

## Data Extraction

Data extraction collects the information needed to describe the studies, assess their quality, and conduct the synthesis. A standardized extraction form should be developed and piloted on two to three included studies before full extraction begins. The form captures study characteriztics, participant or animal characteriztics, intervention details, comparator details, outcomes, and results.

Study characteriztics include author, year, country, study design, and setting. Animal characteriztics include species, breed, sex, age, weight, and health status. For laboratory animal studies, record the strain, model of disease or injury, and induction method. Intervention details include the agent or procedure, dose or intensity, route of administration, frequency, duration, and timing relative to disease onset. Comparator details include the type of control, such as sham, vehicle, no treatment, or active comparator.

Outcome data require particular attention. Record the outcome definition, the time point of measurement, the measurement method, and the results for each group. For continuous outcomes, extract means, standard deviations, and group sizes. For dichotomous outcomes, extract event counts and group sizes. When results are reported graphically only, contact the authors or use data extraction software with appropriate caution. Record any adverse events reported, even when they are not the primary outcome of interest.

Extraction should be performed in duplicate, with disagreements resolved by consensus or third-party adjudication. The extraction form should be made available as supplementary material to the published review.

## Data Synthesis

The synthesis approach depends on the nature of the included studies and the degree of heterogeneity among them. When studies are sufficiently similar in design, intervention, and outcome measurement, meta-analysis may be appropriate. When studies are too heterogeneous, or when data are missing or reported inconsistently, qualitative synthesis is the appropriate alternative.

Meta-analysis of animal studies follows the same statistical principles as meta-analysis of clinical trials, but with additional considerations. Between-study heterogeneity is typically large in animal studies, as demonstrated in the systematic review of mesenchymal stem cell therapy for traumatic brain injury, which reported substantial heterogeneity across twenty-eight controlled studies. Random-effects models are generally preferred because they account for between-study variance. Stratified meta-analysis and meta-regression can explore sources of heterogeneity, such as the timing of treatment, the dose, or the animal model used.

Publication bias assessment uses funnel plots and statistical tests such as Egger's regression test. These methods have limited power when the number of included studies is small, and their interpretation requires caution. The review of dorsal root ganglion stimulation included only fourteen studies and could not perform meta-analysis because of heterogeneity and missing data, yet still provided a structured synthesis of the evidence.

When meta-analysis is not possible, the qualitative synthesis should be systematic and transparent. Group studies by intervention, by animal model, or by outcome domain. Describe the direction and consistency of effects across studies, and identify factors that may explain differences in results. The review of 3D-bioprinted scaffolds for spinal cord injury used qualitative synthesis because the data were not suitable for meta-analysis, yet still reached conclusions about functional recovery across eleven studies.

## Reporting the Review

The final report should follow the [PRISMA 2020 statement](https://www.equator-network.org/) for systematic reviews, adapted for the veterinary context. The [ARRIVE guidelines](https://arriveguidelines.org/) provide the complementary reporting standard for the animal studies included in the review, and reviewers should assess whether included studies adhered to these standards. The report should include the search strategies for each database, the full eligibility criteria, the risk of bias assessments, the extracted data, and the synthesis methods and results.

The discussion section should address the strength of the evidence, the limitations of the included studies, and the implications for clinical practice and future research. The review should state clearly where the evidence base is limited or contested, and where species, production system, or patient status may alter the applicability of the findings.

## Recognized Complications and Failure Modes

Systematic reviews of veterinary therapeutic interventions fail in characteriztic patterns. The most consequential is the mismatch between the review question and the available evidence. When the primary literature consists of small, heterogeneous animal model studies, the pooled effect estimate may be precise but clinically meaningless. This was observed in a systematic review of mesenchymal stem cells for traumatic brain injury, where twenty-eight controlled studies yielded a statistically significant effect but the authors reported modest study quality and large between-study heterogeneity [Peng et al., institutional publication](https://pubmed.ncbi.nlm.nih.gov/25881229/). The review produced a number, but the number could not support a clinical recommendation.

Publication bias is a second major failure mode. Animal studies with null or negative results are less likely to be published, and the review that fails to search grey literature, conference abstracts, and thesis repositories will overestimate treatment effects. Funnel plot asymmetry and Egger's regression provide statistical detection, but these tests require a minimum of ten studies and lose power with fewer. The earlier decision to register the protocol becomes critical here: a registered protocol with a prespecified search strategy allows the reader to distinguish selective reporting from genuine synthesis.

A third failure mode is the conflation of in vitro and in vivo evidence. A systematic review of dorsal root ganglion stimulation for pain included both study types and found that all in vitro studies combined neurostimulation with drugs, while in vivo studies used different stimulation parameters [Vuka et al., institutional publication](https://pubmed.ncbi.nlm.nih.gov/29152818/). The authors could not perform meta-analysis because of heterogeneity and missing data. Pooling these designs without stratification produces an effect estimate that answers no clinically relevant question.

| Observation | Likely cause | Discriminating check |
|---|---|---|
| Pooled effect is large but confidence interval is narrow | Overly homogeneous studies, possible duplicate data | Verify independent study populations, check author networks and institutional overlap |
| Heterogeneity statistic is high but I² is low | Small sample sizes inflate sampling variance | Examine forest plot for outlier studies, consider sensitivity analysis excluding the smallest studies |
| Funnel plot asymmetry detected | Publication bias or true small-study effects | Compare results from published and grey literature, use trim-and-fill only as a sensitivity analysis |
| Meta-analysis impossible due to missing data | Poor reporting of variance measures in primary studies | Contact corresponding authors, consider imputation only if missingness is below 20% and random |
| Effect reverses when a single study is removed | One influential study drives the result | Re-examine that study's risk of bias, report both with and without the study |

## Common Errors and Corrective Actions

Less experienced reviewers frequently confuse study quality with reporting quality. A study that follows the ARRIVE guidelines may still have high risk of bias, and a poorly reported study may have been conducted rigorously [NC3Rs, ARRIVE guidelines](https://arriveguidelines.org/). The risk of bias assessment evaluates what was done, not what was written. The corrective action is to assess each domain independently and to contact authors for clarification before excluding a study on reporting grounds.

A second error is the failure to distinguish between statistical heterogeneity and clinical heterogeneity. Statistical tests quantify variation in effect sizes, but the more important question is whether the studies are asking the same clinical question. A review of curcumin for potentially malignant disorders included randomised trials, observational studies, and case series in a single synthesis [Ara et al., institutional publication](https://pubmed.ncbi.nlm.nih.gov/27072233/). This design mixing produces an average that applies to no specific population. The corrective action is to prespecify the minimum study design for inclusion and to perform subgroup analysis by design if mixing is unavoidable.

A third error is the overinterpretation of subgroup analyzes. Subgroups defined post hoc, without a biological rationale and without adjustment for multiple comparisons, generate hypotheses instead of test them. The corrective action is to limit subgroup analyzes to those prespecified in the protocol and to interpret any post hoc finding as exploratory.

## Limitations of the Current Evidence

The veterinary evidence base for therapeutic interventions is constrained by species diversity, small sample sizes, and the predominance of experimental models over clinical trials. Systematic reviews in this field frequently rely on induced disease models instead of spontaneous disease, and the translation from model to patient remains uncertain. A review of 3D-bioprinted scaffolds for spinal cord injury included eleven studies, all using a rat transection model, and the authors could only synthesise data qualitatively because the data were unsuitable for meta-analysis [Szymoniuk et al., institutional publication](https://pubmed.ncbi.nlm.nih.gov/36858280/). This pattern, qualitative synthesis with explicit acknowledgement of quantitative limitations, is the honest default for much of veterinary therapeutics.

Expert opinion still differs on the role of systematic review in guiding clinical practice. Some argue that the evidence base is too thin to support formal synthesis and that narrative review better serves the practitioner. Others maintain that systematic methods expose the weakness of the evidence and prevent overconfident clinical claims. Both positions have merit, and the reviewer should state clearly where the evidence supports a recommendation, where it supports only a research priority, and where it supports neither.

## Referral, Consultation, and Reporting

The systematic review team should include a librarian or information specialist from the outset, not as a consultant after the search has failed. A statistician with experience in meta-analysis should be involved before data extraction begins, because the choice of effect measure and the handling of missing data are decided at protocol stage. When the review addresses a production animal intervention, a veterinary epidemiologist familiar with herd-level outcomes should be consulted, because the unit of analysis and the clustering of outcomes differ from companion animal studies.

Regulatory reporting obligations arise when the review identifies a previously unrecognised adverse effect of a licensed product. The reviewer should consult the relevant national pharmacovigilance authority and the [AVMA practice resources](https://www.avma.org/resources-tools) for guidance on reporting pathways. For interventions affecting food animals, the [WOAH terrestrial animal health standards](https://www.woah.org/en/what-we-do/standards/codes-and-manuals/terrestrial-code-online-access/) may apply where the intervention influences trade-related disease control. Species-specific pharmacology and withdrawal information should be verified against a current formulary such as the [MSD Veterinary Manual](https://www.msdvetmanual.com/) before any clinical extrapolation is attempted.

## Frequently Asked Questions

### How Much Time and Money Should I Budget for a Veterinary Systematic Review?

A focused veterinary systematic review typically requires six to twelve months of part-time work from a two person team. Costs include database access, reference management software, interlibrary loans, and possibly statistical consultation for meta-analysis. If you lack institutional access to major databases, search Google Scholar and regional veterinary journals, then request full texts directly from authors. The [EQUATOR Network reporting guideline library](https://www.equator-network.org/) can help you identify which reporting standards apply before you begin, saving revision time. If funding is unavailable, restrict your question to a narrow intervention and a single species, which reduces screening volume substantially. A well-scoped review is preferable to an abandoned broad one.

### What Should I Do When the Literature Is Too Heterogeneous for Meta-Analysis?

Heterogeneity is common in veterinary intervention reviews. When statistical pooling is inappropriate, conduct narrative synthesis with structured tables summarizing study characteriztics, effect directions, and risk of bias. The systematic review of dorsal root ganglion stimulation in animal pain models could not perform meta-analysis due to heterogeneity and missing data, yet still produced a useful evidence map [Vuka et al., 2018](https://pubmed.ncbi.nlm.nih.gov/29152818/). Present results by intervention subtype, species, or outcome domain. Use vote counting only as a supplementary approach, never as the primary synthesis, and clearly label its limitations. Consider meta-regression if you have at least ten studies and a priori hypotheses about effect modifiers.

### How Do I Handle Species Differences When Applying Findings Across Animals?

Species differences affect drug metabolism, disease pathophysiology, and outcome measurement. Restrict your review question to one species unless you have a specific rationale for cross-species synthesis. If you include multiple species, pre-specify species as a subgroup analysis and interpret between-species differences cautiously. For example, reviews of mesenchymal stem cell therapy in traumatic brain injury models have shown that biological factors such as cell source and timing influence efficacy, and these factors vary across experimental models [Peng et al., 2015](https://pubmed.ncbi.nlm.nih.gov/25881229/). Consult species-specific clinical references such as the [MSD Veterinary Manual](https://www.msdvetmanual.com/) when assessing whether outcome measures are clinically meaningful across species. Do not assume that a positive result in rodents predicts efficacy in companion animals or livestock.

### What Records Must I Keep During the Review Process?

Maintain a complete audit trail from protocol to publication. Keep the registered protocol with version dates, all search strategies with exact dates and database interfaces, full screening decisions with reasons for exclusion, data extraction forms, and correspondence with study authors. Store these files in a version-controlled repository accessible to both reviewers. The [ARRIVE guidelines](https://arriveguidelines.org/) specify the minimum reporting items for animal studies, and your extraction forms should capture these items systematically. Retain screening logs even for excluded studies, as peer reviewers will request them. If you use spreadsheet software, lock cells and maintain a change log. Your records must allow another team to replicate your search and reproduce your decisions.

### How Should I Explain Review Findings to a Client or Practice Supervisor?

Translate effect sizes into clinically meaningful terms. State the number needed to treat if calculable, the expected improvement in a specific outcome, and the confidence interval around that estimate. Distinguish statistical significance from clinical importance, and acknowledge when the evidence base is too limited for firm recommendations. The systematic review of 3D-bioprinted scaffolds for spinal cord injury, for example, found functional improvement in all included rat studies but could not support meta-analysis, so clinical extrapolation remains uncertain [Szymoniuk et al., 2023](https://pubmed.ncbi.nlm.nih.gov/36858280/). Present the review as one input into clinical decision making alongside patient factors and owner preferences. Offer a written summary with the review citation and a plain language explanation of the certainty of evidence.

### Can I Update an Existing Systematic Review Instead of Starting a New One?

Yes, provided you follow established methods. Contact the original authors to confirm whether an update is in progress. Search for the review in registries and check whether the original protocol specified an update timeline. Re-run the original search from its end date, document any changes to the search strategy, and screen using the same eligibility criteria. If the original review is outdated in methods, you may need to redo risk of bias assessments with current tools. The [WOAH terrestrial animal health standards](https://www.woah.org/en/what-we-do/standards/codes-and-manuals/terrestrial-code-online-access/) illustrate how standards evolve, and the same applies to systematic review methodology. An update that finds no new studies is still publishable as a confirmation of stability, but you must report the search date transparently.

## Related Clinical & Scientific Guides

* [Conducting Systematic Reviews of Veterinary Diagnostic Test Accuracy](/knowledge/veterinary-medicine/veterinary-research-methods/conducting-systematic-reviews-veterinary-diagnostic-test-accuracy)
* [Bias in Veterinary Research: Types, Sources, and Mitigation](/knowledge/veterinary-medicine/veterinary-research-methods/bias-veterinary-research-types-sources-mitigation)
* [Cluster Randomized Trials in Veterinary Research: Design and Analysis](/knowledge/veterinary-medicine/veterinary-research-methods/cluster-randomized-trials-veterinary-research-design-analysis)


## References and Further Reading

- [Systematic review and meta-analysis of efficacy of mesenchymal stem cells on locomotor recovery in animal models of traumatic brain injury.](https://pubmed.ncbi.nlm.nih.gov/25881229/). 2015.
- [Electrical Stimulation of Dorsal Root Ganglion in the Context of Pain: A Systematic Review of In Vitro and In Vivo Animal Model Studies.](https://pubmed.ncbi.nlm.nih.gov/29152818/). 2018.
- [Exploring the logic and conducting a comprehensive evaluation of AdipoRon-based adiponectin replacement therapy against hormone-related cancers-a systematic review.](https://pubmed.ncbi.nlm.nih.gov/37864589/). 2024.
- [GABA(B) receptor agonists for the treatment of drug addiction: a review of recent findings.](https://pubmed.ncbi.nlm.nih.gov/11841892/). 2002.
- [Research on curcumin: A meta-analysis of potentially malignant disorders.](https://pubmed.ncbi.nlm.nih.gov/27072233/). 2016.
- [The application of 3D-bioprinted scaffolds for neuronal regeneration after traumatic spinal cord injury - A systematic review of preclinical in vivo studies.](https://pubmed.ncbi.nlm.nih.gov/36858280/). 2023.
- [ARRIVE Guidelines 2.0 for Reporting Animal Research](https://arriveguidelines.org/). PLOS Biology, 2020.
- [EQUATOR Network Reporting Guidelines](https://www.equator-network.org/). EQUATOR Network.
- [MSD Veterinary Manual, Professional Edition](https://www.msdvetmanual.com/). MSD Veterinary Manual.

## Related Articles

- [Conducting Systematic Reviews of Veterinary Diagnostic Test Accuracy](/knowledge/veterinary-medicine/veterinary-research-methods/conducting-systematic-reviews-veterinary-diagnostic-test-accuracy)
- [Narrative Reviews in Veterinary Medicine: When and How to Write Them](/knowledge/veterinary-medicine/veterinary-research-methods/narrative-reviews-veterinary-medicine-when-how)
- [Conducting Pharmacovigilance Studies in Veterinary Medicine](/knowledge/veterinary-medicine/veterinary-research-methods/conducting-pharmacovigilance-studies-veterinary-medicine)
- [Conducting Qualitative Research in Veterinary Settings](/knowledge/veterinary-medicine/veterinary-research-methods/conducting-qualitative-research-veterinary-settings)
- [Performing Economic Evaluations of Veterinary Interventions](/knowledge/veterinary-medicine/veterinary-research-methods/performing-economic-evaluations-veterinary-interventions)

> This article is educational professional reference material for veterinary audiences. It is not a substitute for veterinary diagnosis, individual clinical judgment, current product labeling, or applicable regulatory requirements.