Background
The placebo effect is the measureable or observable response in the patient’s health or behavior that can not be attributed to the administration of a medication or a valid treatment plan. Placebo is derived from a Latin word meaning “I shall please” and is defined as a pharmacologically inert substance; an example of a placebo is a saline solution. When this substance is given to an individual if they experience an effect that would be comparable to a pharmacological substance then it can be stated that they are experiencing the placebo effect.
References to the placebo effect can be found in medical journals dating back to the early 1950’s. The research article “The Powerful Placebo” published in 1955 found that 35% of 1,182 patients reported relief of their symptoms with the application of a placebo drug (Perry, S. W. & Heidrich, G., April 1981). Due to this research additional data has been collected which indicate that anywhere between 50 and 60% of subjects will respond to a placebo if they are experiencing specific medical conditions such as depression, pain and stomach complaints (Perry, S. W. & Heidrich, G., April 1981).
As more research is done on how the brain influences the body the power of suggestion and the placebo effect have become a more integral part of the research process (Perry, S. W. & Heidrich, G., April 1981). It is believed that the placebo works by acting on the emotional impact of an idea which is transmitted to the limbic system. Once the limbic system is involved the brain can utilizing a biochemical network resulting in the patient experiencing either symptoms of an illness or the relief of the symptoms (Perry, S. W. & Heidrich, G., April 1981).
The placebo has been used throughout medical research as a way that the effectiveness of a new medication or the psychological effects are being studied. This requires deception on the point of the researchers because research has shown that when a patient knows that the medication being administered is a placebo the effects of the drug will be minimized (Connelly, R. J., 1987). While most drugs do not experience a problem when compared to a placebo it has happened in enough medication trials that researches are starting research into the effects that a placebo can have on a patients or subjects health (Enserink, M., 1999).
Research is not the only area where placebos are utilized. In some situations doctors will prescribe placeboes for pain management. The use of placeboes in this situation is to prevent drug seeking behavior from addicts (Pasero, C. L., January 1998). Following the research the best way to ensure that the placebos work is to provide misinformation to the patient. In the research setting the misinformation provides valid data, however, used in the medical field as a tool for pain management raises ethical issues.
As research is reliant upon the placebo to provide validation of a drugs effectiveness studies in which the placebo receives better results then the medication are troubling for the drug companies and the researchers. A promising new treatment plan can be derailed by the placebo effect (Enserink, M., 1999).
Mass Psychogenic Illness (MPI) is the collective occurrence of self-reported physical symptoms without the presence of an identifiable pathogen (Lorber, W., Mazzone, G. & Kirsch, I., 2007). Through the increased risk of biochemical terrorist attacks and the transmissions of new diseases an environment is created in which psychogenic symptoms are more likely to occur. This could be devastating to public health officials who are overwhelmed by patients experience symptoms especially if this occurs after a biochemical attack.
Aims of the Proposed Research
As Beecher’s study created an outpouring of research examining the effects a placebo can have on a patients symptoms, this study plans on conducting an experiment that has been done with slight modifications in an effort to eliminate one of the flaws in the research. In order to compare the results of a study to the widest possible population the population of the study must represent the general population.
The data gathered originally by Illness by Suggestion: Expectancy, Modeling, and Gender in the Production of Psychosomatic Symptoms written by William Lorber, Giuliana Mazzone, and Irving Kirsch cannot be compared to the general population because only females participated in the study. Since men were not involved in the study, the effects of suggestion combined with the administration of a perceived toxin can not be correlated.
In this study the researchers examined a group of undergraduate females in a study that was focused on determining how an individual who expected a specific situation and when that situation is combined with individuals modeling behaviors consistent with the perceived notion would respond in a predictable fashion (Lorber, W., Mazzone, G. & Kirsch, I., 2007). This response would be based on the suggestion of an illness rather then an actual illness. This study while it was well designed and well executed only tested a female population, it is important that this study be conducted again ensuring that males are included in the study not only as participants but as confederates. By doing this the data collected will show a greater representation of the actual population and be more statistically relevant.
Method
Participants
This study will gather its participants from an introductory undergraduate psychology course. The students will be asked to sign up for the experiment by their professor. The results of the study would be more reliable if 150 individuals participated in the study. In order to obtain participants from both the male and female sex 75 slots will be held for members of each sex, once those slots are full, individuals who wish to participate would be placed on a waiting list. Individuals who choose to participate who have asthma or other respiratory conditions will be excluded from the study.
Design
This study will use a 2 x 2 x 2 mixed-model experimental design where two between-subject factors such as inhalation and observation and one within-subject-factor such as specified symptoms vs. other symptoms (Lorber, W., Mazzone, G. & Kirsch, I., 2007). The participants will be randomized to either inhale or not inhale a substance that the researchers described as a possible environmental toxin reported to produce several symptoms including headache, nausea, itch skin, and drowsiness. The actual substance will be odorless air. The participants will also be randomized to either observe or not observe a member of the research team posing as a participant also known as a confederate inhale the substance and exhibit the symptoms (Lorber, W., Mazzone, G. & Kirsch, I., 2007).
At ten minute intervals the participant and confederate will asked to verbal rate four specified symptoms such as the headache, nausea, itchy skin and drowsiness as well as four additional symptoms such as watery eyes, tightness in chest, breathing difficulty and a scratch throat (Lorber, W., Mazzone, G. & Kirsch, I.,2007). The specified symptoms were chosen specifically because they are similar to those symptoms of MPI 17-19. The four additional symptoms were chosen because there presence could be observed by the participant’s behavior (Lorber, W., Mazzone, G. & Kirsch, I., 2007).
The participants will be debriefed after the study is concluded.
Measures and Materials
Metered-dose inhalers with no additives will be necessary for this study. This will allow the researches to provide the initial stimulus and ensure that the participants are not accidently exposed to harmful chemicals.
Procedures
This experiment will be conducted at the universities student health center. When the students signed up for the experiment they were asked to choose a specific date and time. They will also be told that the experiment will be conducted with two participants each session and each session will be videotaped. One of the participants will be a member of the research team and assume the role of the confederate. The pairings will be set up as male x male (confederate), female x male (confederate), male x female (confederate), and female x female (confederate). These paring will allow the researchers to examine the interactions between the confederate and the participants as it relates to increasing or decreasing the power of suggestion. The students will be given the following instructions by the experimenter:
I am a research assistant and part of the research team that is studying a suspected environmental toxin. This substance has been reported to produce a number of temporary symptoms in several workplaces. The most frequently reported symptoms that have been reported include headaches, nausea, drowsiness and itchy skin. These symptoms seem to develop quickly but are relatively mild and do not last long. This study is being conducted so that accurate observation of the symptoms produced by this substance can be accurately gathered.
The research team is also attempting to better understand the progression of the illness. To better conduct the research we are asking several participants to inhale the possible toxin. Those participants who do not inhale the substance will be placed in the control group. The control group is important because it will allow the research team to control for symptoms that do not result from the toxin. A coin toss will decide which group you will be placed in (Lorber, W., Mazzone, G. & Kirsch, I., 2007).
At this point the participants will be offered a chance to remove themselves from the experiment and those who chose to continue with the experiment will be given an Institutional Review Board approved informed consent form.
Baseline symptoms for each of the eight symptoms will be obtained and then depending on the results of the coin toss will be asked to inhale from the placebo metered-dose inhaler. After the inhalation they will be required to hold their breath for three seconds so that the exposure time will be standardized (Lorber, W., Mazzone, G. & Kirsch, I., 2007). Symptoms will then be evaluated every ten minutes during the hour. The participants will be allowed to read magazines or work on homework in between evaluations, however, the confederate will interject specific comments during the hour long experiment.
In the observe symptoms condition the confederate will model the four specified symptoms by displaying overt behaviors and statements describing how the individual felt. The displayed symptoms will be increased over the first thirty minutes of the experiment and then slowly decrease (Lorber, W., Mazzone, G. & Kirsch, I., 2007). In the not-observe condition the confederates will not display any symptomatic behavior. In both conditions the confederate will be instructed not to begin any conversations with the participant and while responses to participant initiated conversation will be allowed they will try to be kept short (Lorber, W., Mazzone, G. & Kirsch, I.,2007).
The sessions will be videotaped to assist in the assessment between the behavior and the changes in self-reported symptoms. Two research assistants will view the videotapes at a later date so that independent verification will occur.
Data Analysis
The researchers in the original study used the Analysis of Covariance (ANCOVA) test when examining the results. This test is similar to the methods used in the analysis of variance. It functions by including the covariates into the model. This is important in this experiment because it allows the researchers to account for the variable created by the confederate as well as measuring the placebo effect on the subject’s behavior.
This test will also allow the researchers to evaluate which combination of sex of confederate vs. sex of the subject. The sex of the confederate is the covariate because it is not related to the administration of the substance but can affect the subject’s response.
If the covariate can be associated with the effect of treatment then analyzing the covariance may be of more use then analyzing the variance.
It is believed that this research will provide different results then the original research through the addition of confederates including male x male (confederate), male x female (confederate), female x female (confederate) and female x male (confederate).
In the original experiment the participants who inhaled the placebo showed an increase in general symptoms and the specified symptoms showed significant increases above the increase in the general symptoms (Lorber, W., Mazzone, G. & Kirsch, I., 2007).
The research will also be analyzed using a mixed-model ANOVA. Using this statistical analysis proved the significant effect was limited to those participants who inhaled the placebo (Lorber, W., Mazzone, G. & Kirsch, I., 2007).
The significant event in the first experiment was the significant Participant Inhalation x Symptom Type interactions in which the participants who inhaled the placebo reported greater increase in the specified symptoms as well as the general symptoms in comparison with the participants who did not inhale the placebo experienced no significant difference between the specified and general symptoms (Lorber, W., Mazzone, G. & Kirsch, I., 2007).
Interpretation of Anticipated Results
With the addition of mixed sex confederates it is expected that an increase in the two way interactions between the gender and the experienced symptoms will occur and that the placebo effect will be greater when the confederate is of the opposite sex.
The four groups of data that are being compared are:
- female x female (confederate).
- female x male (confederate).
- male x male (confederate).
- male x female (confederate).
The ANCOVA test will determine which of the covariates produced the greatest placebo effect in the subjects.
As an individual beliefs and expectations about a situation combined with how much of an effect suggestion has on the individual will have a biochemical effect resulting in a response by the bodies neuochemical system. This is consistent with current medical opinion that a patient’s attitude and beliefs will effect their physical well-being.
This study focused on behaviors that would stimulate the psychogenic symptoms in the participants. The manipulation of the expectations and environment combined with the desired behaviors being modeled by a confederate created a situation in which the placebo effect would appear. By manipulating the expectations of the participants who were expecting the inhaled substance to be a toxin is consistent with the response expectancy theory as well as contradicting the theories that predict nonspecific effects. The researchers are expecting that the participants who believe that they are inhaling the toxin will exhibit the same symptoms described to them before the experiment began. This will happen because of the manipulation of the environment and there expectations.
References
Connelly, R. J. (1987). Deception and the Placebo Effect in Biomedical Research. Ethics and Human Research, 9(4), 5-7.
Enserink, M. (1999). Can the Placebo be the Cure? Science, 284(5412), 238-240.
Lorber, W., Mazzone, G. & Kirsch, I. (2007). Illness by Suggestion: Expectancy, Modeling, and Gender in the Production of Psychosomatic Symptoms. The Society of Behavioral Medicine, 33(1), 112-116.
Pasero, C. L. (1998). Pain Control: When the Physician Prescribes a Placebo. The American Journal of Nursing, 98(1), 52-53.
Perry, S. W. & Heidrich, G. (1981). Placebo Response: Myth and Matter. The American Journal of Nursing, 81(4), 720-725.