This article is the first in a series about “Clinical Research”. After this initial (general) introduction, further articles will guide you through the process of carrying out a clinical research study, from idea conception through protocol development, through statistical analysis of the results, and reporting in the literature.
The actual term “Clinical Research”, is further categorised into “Pre-clinical” and actual “Clinical”. Pre-clinical research includes such evaluations as physical properties testing, bond strength to tooth structure testing, wear testing, and handling properties. Pre-clinical testing is usually carried out in Biomaterials laboratories of dental schools. Although there are 55 dental schools in the U.S. and 10 dental schools in Canada, generally only a few are well suited to perform the required pre-clinical research by virtue of their expertise and their laboratory facilities. Pre-clinical research always requires the careful preparation of laboratory “specimens” in order to carry out the experiments, which are always outlined within a clearly defined protocol. The preparation of specimens is a skill, which requires significant experience, and to which we must pay close attention in order to be comfortable with the validity of the outcome.
Actual “Clinical” research involves human subjects, and is generally preceded by pre-clinical research in order to establish some of the “outcome measures” or “endpoints” of the planned clinical investigation. Many different types of studies involving patients can be performed, which are categorised based on their study design. The design of the clinical investigation is critical to the ultimate meaning of the results, and will be discussed in detail in the next article in this column entitled “Clinical Research Design and the Meaning of its Results”.
Clinical Research Design And The Meaning Of Its Results
The following is Part II in a series of articles regarding the process of evaluating products in an effort to gain valuable information for the practitioner This piece reviews the design of a study, and the way in which that design can impact the meaning of results The design of any clinical or pre-clinical evaluation is the critical component in the determination of the meaning of the results. A poorly designed study will provide data, but one must question the value of the data. The design of the study has several elements; all of which must have significant attention paid to them.
First the determination must be made as to what is being evaluated. In the case of a new restorative material, the material in question is being evaluated and subsequently, “the endpoints”, or the actual parameters to be evaluated must be determined. For example, if one is looking at the wear characteristics of a certain material, we often look at microns loss of substance on the occlusal surface within a simulated wear environment in vitro, in the laboratory.
In order to adequately assess the meaning of the results, we must have a counterpart specimen, which is considered the control compared with the experimental specimen being evaluated. This control will provide a second (related) material which is simultaneously subjected to the same conditions to which the experimental specimen is subjected, and which will provide a meaningful counterpart to compare. In order for data to be meaningful, specimens must be randomly selected and randomly assigned into experimental and control categories. In other words, if experimental and control specimens are being compared, each must have an equal chance of selection.
If one compares a new restorative material to an existing one, then only after the informed consent is obtained from the patient, can a selection be made as to which restorative material; i.e., the control or experimental, is to be placed within that particular patient. In other words, the patient must give consent to give either the experimental (usually newer material) or control (usually existing material) placed as a restoration in their mouth. If the consent is obtained after the determination of what restoration is to be placed, then this is not a random assignment, and creates significant bias in the experiment. There are many other factors related to selection criteria, including “inclusion criteria and exclusion criteria”. These generally have to do with age of the patient, gender of the patient, medical history, type of teeth, and number of teeth present in the mouth, type of occlusion, and a variety of conditions which collectively can significantly affect the outcome of the experiment.
It must be kept in mind that any scientific experiment looks at a random sample from the population, and therefore, we must carefully select the patients so they are representative of the population as a whole, otherwise the results are not meaningful at all. The determination of what statistical evaluations will be made on the data, generally comparing experimental and control specimens must be determined at the time of design of the experiment. Most critical to statistical analysis is the determination of what statistical test is appropriate. We will not discuss statistical analyses at this point, but in a future piece, we will discuss some rudimentary elements and statistical analyses of clinical experiments.
Therefore, the next time you read a scientific article, it is important to carefully look at the premise of the experiment, the purpose of the experiment, and the way in which the design of the experiment was established. The design will determine the value of the results. Future articles here will discuss some of the other elements of clinical research design, including obtaining informed consent from the patient, the meaning of informed consent, as well as other types of experiments which are not patient- connected, and how they are established.
.
