Medical research is behind virtually every advance we see in hospitals and healthcare centers: from a vaccine to a new surgical technique or an app that remotely monitors a chronically ill patient. Without this silent work carried out in laboratories, clinics, and databases around the world, we would still be without answers for many infectious diseases, tumors, or chronic conditions that we can now diagnose better and treat more effectively.
At the same time, the avalanche of clinical studies, news reports, and "major discoveries" that appear daily can be overwhelming. It's not always easy to understand what kind of research lies behind a headline, how reliable it is, or what real implications it has for an individual's health. That's why it's helpful to know what medical research is, its main types, real-world examples, and what questions to ask before accepting a scientific breakthrough as valid.
What is medical research and why is it so crucial to health?
When we talk about medical or biomedical research, we are referring to the body of scientific studies aimed at understanding how the human body works, why we get sick, how to better diagnose, treat more safely and effectively, and prevent health problems. This includes everything from basic laboratory experimentation to studies with patients, as well as the analysis of large databases and public health projects.
Within this broad framework, clinical research occupies a key place: it is the part of medical research in which people (patients or healthy volunteers) participate directly, without separating them from their personal data, images, blood or tissue samples, and medical history. The goal is to understand disease and health in all their complexity, not only at the cellular or genetic level, but also within the real-world context in which the person lives.
Furthermore, the concept of translational research has gained traction , acting as a bridge between laboratory findings and clinical practice. This approach combines basic research techniques (such as the study of genetic markers) with patient data, but without the same level of integration as in strictly clinical research. A typical example is the analysis of genes that are subsequently associated with poorer survival in certain tumors, paving the way for targeted therapies.
In practice, medical research allows for the discovery of new methods for detection, diagnosis, treatment, prognosis, and prevention . Clinical trials of new drugs, research on the natural progression of chronic diseases, and studies of risk factors affecting a community are all pieces of the same puzzle that ultimately influence clinical practice guidelines and the daily decisions of healthcare professionals.
Above all, research involving human subjects demands meticulous attention to ethics and safety : informed consent, review by ethics committees, monitoring for adverse effects, transparency in results, and a clear social value. The aim is for the benefits of the studies to clearly outweigh the risks and for participants to ideally gain some direct advantage.
Basic, clinical and translational research: how they connect
The biomedical research process is often described as a step-by-step sequence. At one end is basic research, focused on fundamental biological mechanisms: laboratory studies (in vitro) and experiments in non-human animals (in vivo). These initial steps are essential to demonstrate a certain level of safety and efficacy before considering testing an intervention in humans.
When an idea or product passes these initial tests, clinical research comes into play . Here, work is done with human subjects, and the focus shifts to safety and efficacy under conditions more similar to real-world scenarios. It is the final link before something reaches routine clinical practice, and also the final filter: if a basic or translational finding is not confirmed in patients, it will have no real impact on daily medicine.
In between, there is translational research , which aims to ensure that what is discovered in the laboratory doesn't just sit "on the shelf." For example, a group might identify a genetic biomarker in tumor samples and then study in patients whether that biomarker is associated with a worse prognosis or with a response to a specific treatment. This integration of both worlds requires mixed teams with both clinical and basic training.
For this entire system to function, it is essential that research questions not be determined solely by the availability of flashy technology in the laboratory. Good clinical research begins with real patient needs: what problems remain unresolved, what diagnoses raise doubts, and what current treatments could be improved. Focusing only on what “can be measured” without assessing clinical relevance can lead to unhelpful results.
Another key aspect is distinguishing between statistical significance and clinical significance . A study may find numerically significant differences (for example, a minimal decrease in a lab value), but these may not change the patient's life or prognosis. The challenge for current research is to design studies focused on outcomes that are truly important for those suffering from the disease.
Types of clinical research according to their objective
Clinical research can be classified according to the question it seeks to answer . Although in practice a study may overlap with several categories, this classification helps to better understand the purpose of each study:
- Case-based or descriptive research: focuses on describing what patients and diseases are like in clinical and epidemiological terms (who gets sick, with what characteristics, at what time, etc.).
- Etiological investigation: seeks the causes or factors that are associated with the development of a disease, usually through analytical observational studies or explanatory pathophysiological work.
- Diagnostic research: evaluates the real value of clinical or complementary tests to detect a disease (sensitivity, specificity, predictive values).
- Therapeutic research: analyzes the efficacy and safety of biomedical treatments, procedures, and devices.
- Prognostic researchIt studies the natural history of the disease, the evolution of patients, and the factors that influence the short- and long-term outcome.
In all cases, it is essential that the research team has a solid background in clinical practice, research methodology, and epidemiology . Only then can they choose the appropriate design for the research question, clearly define the outcomes to be measured, and assess whether the findings truly have relevance to patients' lives.
Clinical study designs: observational and experimental
A key decision when starting any project is choosing the study design . This involves assessing the available information, practical feasibility, the required sample size, ethical considerations, costs, and, above all, whether the researcher will simply observe or intervene.
Based on this role, two main groups are distinguished: observational studies (the researcher does not modify reality, but only records and analyzes it) and experimental studies (an intervention is introduced and compared with another intervention or a control group). Each type of design has strengths and limitations, and provides different levels of evidence depending on the health problem being addressed.
Observational studies: description and analysis of reality
Observational studies are used to describe diseases, explore associations between risk factors and outcomes, or estimate the frequency of a problem in a population. They can be descriptive (simply reporting what happens in a group, without a comparison group) or analytical (comparing different groups to look for associations).
Within the observational studies, we find several fundamental designs:
Case reports and case series
Case reports describe one or a few patients in detail (usually up to 10), while case series include a larger number. They lack a control group and are very common in scientific journals because they allow for the communication of striking findings: very rare diseases, atypical presentations, unusual responses to treatment, or novel therapeutic modifications.
For example, several cases of hepatic hydatidosis treated with a specific laparoscopic technique could be published , detailing morbidity, hospital stay, and recurrence. Or dozens of patients with unusual locations of hydatid cysts (peritoneum, retroperitoneum, pelvis, etc.) could be described, analyzing the need for multiple surgeries and the complications.
Their main value is to generate hypotheses and alert about emerging phenomena , although on their own they do not allow establishing causal relationships or estimating risks.
Cross-sectional studies
Cross-sectional studies measure all variables at a single point in time , without follow-up. A classic use is to estimate the prevalence of a disease in a specific population. For example, administering a validated questionnaire to a sample of urban adults to calculate the proportion of people with gastroesophageal reflux disease over a given period.
This design allows for the exploration of associations between factors (such as habits or demographic characteristics) and the presence of disease, although it is difficult to determine whether the factor preceded the health problem or is a consequence of it. These studies are particularly useful when exposure factors do not change over time (sex, blood type), where associations can be interpreted with greater confidence.
Population, correlational, or ecological studies
Population or ecological studies use aggregated population data (rather than individual data) to compare disease or mortality rates between groups or time periods. An example would be analyzing suicide mortality in a country over several years, comparing regions, sex, and age groups, and calculating the rate per 100,000 inhabitants.
These designs allow for the exploration of trends and potential associations on a large scale , although caution is advised: conclusions about individuals drawn from aggregated data can lead to errors (ecological fallacy). Even so, they are very useful in public health and resource planning.
Case-control studies
In a case-control study , the starting point is the outcome: a group of people who already have the disease or event of interest (cases) is selected and compared with another group who do not have it (controls). Then, the study looks back to see how frequently each group was exposed to a potential risk factor.
These studies are typically retrospective and highly effective for investigating rare diseases or those with long latency periods. However, the quality of the study depends heavily on how the cases and controls are defined and selected. Controls should resemble the cases in every way, except that they do not have the disease; they can be drawn from hospitals, the general population, or the close contacts of the cases (family members, neighbors).
An example would be comparing patients with primary inguinal hernias to individuals without hernias who underwent surgery for other reasons, studying the expression of certain proteins involved in collagen degradation (such as MMP2 and TIMP2) to assess whether there is a difference between the two groups. If no significant differences are found, it is concluded that this alteration is not associated with the development of the hernia in that context.
Cohort studies
In cohort studies, a group of people free of the disease at the start is selected, classified according to the presence or absence of an exposure (e.g., substance use, occupational exposure, or a clinical characteristic), and followed over time to see who develops the event of interest.
This design allows for calculating disease incidence and estimating relative risks. Cohorts can be prospective (following individuals into the future) or retrospective (reconstructing follow-up data from existing records). There are also bidirectional cohorts and nested case-control designs within a cohort, which combine advantages of both approaches.
A prospective example would be to follow patients operated on for hepatic hydatid disease , distinguishing between those with intracystic biliary communications and those without, to analyze whether the presence of two or more communications increases the risk of postoperative morbidity. Another example is a case-control study within a cohort of patients with operated hydatid disease to identify which factors (age, previous surgery, laboratory parameters, etc.) are associated with increased morbidity.
Systematic reviews and meta-analysis
Systematic reviews ( SRs) are a special type of observational research in which the "population" consists of previously published studies. The goal is to comprehensively gather all relevant evidence on a specific question, assess the quality of the studies, synthesize their results, and often integrate them quantitatively through meta-analysis.
To achieve this, a rigorous literature search is essential , along with the application of clear inclusion and exclusion criteria, an assessment of the design and methodological quality of each study, and an analysis of potential heterogeneity among them. Combining small studies through meta-analysis increases the overall sample size and statistical power, allowing for more precise estimates.
A classic example is a systematic review comparing the use of opioid analgesics versus placebo in the management of acute abdominal pain, to see if opioid use interferes with diagnosis. After screening hundreds of articles, only the relevant randomized controlled trials are selected, their data are extracted, and combined. If the meta-analysis shows that opioids do not increase the risk of diagnostic error and that they improve patient comfort without delaying decisions, a robust conclusion is reached with a direct impact on clinical practice.
In recent years, network meta-analysis has gained relevance , allowing for the simultaneous comparison of multiple interventions, even when they have not all been directly compared in clinical trials. This method combines direct and indirect comparisons, provided that the primary studies are methodologically comparable, there is low heterogeneity, and assumptions such as transitivity and consistency are met.
Experimental studies: clinical trials and quasi-experiments
Experimental studies are characterized by the researcher introducing an intervention (a drug, a technique, a program) and comparing its effects with those of another intervention or a control (placebo or standard treatment). They are usually prospective and are the primary tool for assessing the efficacy and safety of new therapeutic options.
Clinical trials
The clinical trial is the quintessential experimental design in human medicine. It involves assigning participants to different groups that will receive one intervention or another, generally through random assignment. The aim is to ensure that the groups are comparable in all factors, known and unknown, except for the treatment being studied.
Basic methodological components include randomization sequence generation and concealment , blinding (ideally double-blind: neither patients nor researchers know who receives whom), sample size calculation, and the use of intention-to-treat analysis. Trials can pursue diverse objectives: comparing efficacy, studying dosage, evaluating bioequivalence, or exploring the dose-response relationship.
Clinical trials are divided into two main types: classic explanatory trials , conducted under highly controlled conditions with strict inclusion criteria, and pragmatic trials , designed to determine how an intervention works in the "real world" (general hospital population, emergency departments, primary care). In recent years, randomized controlled trials based on clinical registries have emerged, using data from clinical information systems to recruit and follow large samples at a lower cost, although they require careful monitoring of data quality.
A practical example would be comparing laparoscopic cholecystectomy versus open cholecystectomy through a small incision in patients scheduled for gallbladder surgery. Patients are randomly assigned to both techniques, and variables such as quality of life, operative time, pain, complications, and hospital stay are analyzed. If, for example, a slight initial advantage in quality of life is observed for one group but without significant differences in complications, this information will be incorporated into treatment guidelines.
Quasi-experimental studies
Quasi-experimental studies are designed when random assignment of the intervention is not possible or ethical. They share with clinical trials the introduction of a measure (a drug, a program, an organizational change), but the allocation of who receives it and who does not is determined by external circumstances or clinical decisions, not by chance.
They are simpler and cheaper to conduct, and are often the only alternative when the goal is to evaluate the impact of a health policy or a change in practice under real-world conditions. Their drawback is their greater susceptibility to bias, especially selection and confounding bias, so their interpretation requires caution.
Examples include studies in which preoperative albendazole is administered to a group of patients with hepatic hydatidosis and the concentration of the drug inside the cysts and the viability of the scolices are measured, without random assignment; or repeated-measurement investigations in patients undergoing liver resections, in which the evolution of analytical parameters (liver enzymes, coagulation, blood cell count) is monitored from before surgery until several days later, to describe the pathophysiological changes of ischemia and reperfusion.
Phases of clinical trials and official registries
Before a new drug or technique reaches the market, it must go through several phases of clinical trials . In phases I, II, and III, information is gathered on safety, dosage, efficacy compared to existing treatments, and side effect profile in different patient groups. After approval, phase IV continues to study long-term effects and effectiveness under typical use conditions.
All clinical trials must comply with international and national ethical standards, follow detailed protocols, have the approval of competent authorities, and be registered on public platforms. In Spain, there is the Spanish Clinical Trials Registry (REec) , managed by the Spanish Agency for Medicines and Health Products (AEMPS). Following the COVID-19 pandemic, exceptional measures were introduced to facilitate and expedite the management of trials against SARS-CoV-2 infection.
In addition to the REec, there are other recognized registries such as those of the World Health Organization , the European registry, and the US registry. Registering and publishing trials in these systems contributes to transparency, allows researchers to know which studies are underway, and helps avoid unnecessary duplication of research.
How to read and evaluate a news story about medical research
Headlines about supposed medical breakthroughs based on recent studies appear almost daily . However, not all research carries the same weight, and the media doesn't always provide adequate context. It's wise to maintain a degree of constructive skepticism and ask some questions before drawing conclusions.
When evaluating a study, it is useful to consider aspects such as:
- Was it performed on animals or humans? What works in animal models does not always translate into the same effect in people.
- Who were the participants? Age, sex, ethnic origin, comorbidities… If the population does not resemble the reader or the patients they are interested in, the applicability may be limited.
- What was the sample size? Very small studies may produce striking results by chance; larger studies are needed to confirm them.
- How long did the follow-up last? Especially with medications, sufficient time is needed to detect long-term benefits and risks.
- What type of design was used? Randomized clinical trials usually provide stronger evidence on the efficacy and safety of treatments than observational studies.
- Where was the investigation conducted? The health and social context influences the generalizability of the results.
- What side effects were found? It is just as important to know the risks as the benefits, and to know how often they occur.
- How does the study fit with previous evidence? If it contradicts many previous studies, perhaps more research is needed before changing protocols.
- Who funded the study and who is communicating it? Funding sources and potential conflicts of interest can introduce bias.
Keeping these questions in mind helps to interpret more calmly the "great discoveries" that appear in the press and to have more informed conversations with healthcare professionals about what a new study really means for individual health.
Medical research, public health and global threats
Research is not limited to individual treatments: it is essential for addressing public health challenges . Diseases of poverty, increasing chronic illnesses, maternal mortality, mental health problems, and antimicrobial resistance all require solid data to design effective policies.
Public-private partnerships and innovative funding mechanisms have also emerged, focusing on neglected diseases where the market alone does not provide sufficient incentives for research. These efforts have led to the development of new medicines, vaccines, and diagnostic tools that would otherwise likely not exist.
In parallel, the big data and artificial intelligence revolution (supported by supercomputers ) has ushered in a new era: the massive analysis of clinical, genomic, and even wearable device data allows for the identification of patterns, the anticipation of outbreaks, the personalization of treatments, and the optimization of healthcare resources. However, this potential can only be realized with secure infrastructure, robust software, and a clear ethical framework for the use of information.
Tools for evaluating the quality of studies
Given the enormous amount of published scientific literature, checklists and guides have been developed to help improve the quality of research reports and better evaluate published studies.
Among the most notable are:
- MInCir-EOD: checklist for reporting descriptive observational studies, with 19 items grouped into 4 domains, aimed at ensuring that key information is not missing from the manuscripts.
- STROBE: 22-point guide for the presentation of cross-sectional, case-control and cohort studies, covering title, abstract, introduction, methods, results and discussion.
- MOOSE: specific proposal for the reporting of meta-analyses of observational studies in epidemiology, with emphasis on the search strategy, methods and synthesis of results.
- MInCir-therapy: scale that helps to assess the methodological quality of therapeutic studies, including items on type of design, population size and methodology.
- CONSORT: declaration for the transparent communication of randomized clinical trials, with 22 items and several specific extensions (e.g., for patient-reported outcomes or complex interventions).
- TREND: guide designed for intervention studies without random assignment, useful in many quasi-experiments.
- PRISMA: reference standard for reporting systematic reviews and meta-analyses, with 27 items ranging from title to funding.
These tools are not used to assign "grades" to articles, but to promote greater transparency and rigor , making it easier for readers, reviewers, and editors to know exactly what was done, how, and with what limitations.
Challenges and future of medical research
Despite the increasing demand for scientific evidence to guide clinical practice , conducting research is, paradoxically, more complex than ever. Regulatory requirements, competition for funding, the pressure of patient care on clinicians, and the need for multidisciplinary teams mean that carrying out a successful study is far from easy.
Added to this are the ethical dilemmas linked to new technologies such as gene editing, regenerative medicine , and the massive use of personal data. Guaranteeing respect for individual autonomy, privacy, and a fair distribution of the benefits of research is just as important as achieving spectacular results in the laboratory.
Furthermore, research must always have a clear social value : improving healthcare, patient safety, or the efficiency of the system in some way. Innovation for innovation's sake, disconnected from the real problems of the population, risks consuming resources without providing concrete solutions.
Looking ahead, we expect to see significant advances in the next decade thanks to the combination of genomics, artificial intelligence, nanotechnology , and personalized medicine. Curing currently incurable diseases, designing drugs with algorithms, fully digitizing healthcare systems, and developing new vaccines in record time are becoming less and less utopian goals.
In this context, medical research will remain the essential lever for transforming healthcare : from analyzing the extracellular matrix of a hernia to evaluating a national vaccination policy, and including the design of clinical trials for precision cancer therapies . The major challenge will be maintaining methodological quality, ethics, and a patient-centered approach amidst a rapidly evolving technological and social environment.
