Observational vs. Experimental Study

Please discuss a Method Section from a peer-reviewed journal article, including a link to a journal article, and determine whether the study qualifies as an observational or an experimental study. Due date Feb 24th 2014. Please keep your post as a “Reply” to this message.

20 thoughts on “Observational vs. Experimental Study

    • This study examines the association between type 2 psoriasis and type 2 diabetes and qualifies as an observational study because it utilizes three cohorts of which each was followed for at least ten years. Previous cohort studies have failed to determine the association between psoriasis and the risk of type 2 diabetes (T2D) because none of these studies have adjusted for T2D’s major risk factors, including obesity. Using a retrospective-prospective cohort study, Li et al. (2012) analyzed the relationship between psoriasis and the risk of acquiring T2D in the United States after multivariate adjustment. Their findings highlight a non-significant trend to increased risk for T2D; specifically, they found that younger people (less than 60 years old) developing psoriasis had an increased risk of T2D. Specifically, they used three different cohorts of health professionals (younger women, older women, and older men aged 40-80 years) were followed for ~10 years and evaluated using baseline and biennial questionnaires. Their questionnaire was very integral to the experiment and exhibited a great response rate and confirmation rate (87, 92%). Since diagnosis was based on recall, they utilized a population more likely to have higher recall and regular access to medical records (health professionals).They adjusted relative risks based on age and multiple variables including BMI, alcohol intake, physical activity, smoking, and other risk factors. They gave the predicted relative risk and 95% confidence intervals for each cohort to assess T2D risk for psoriasis and non-psoriasis patients.The article specifically looked at women younger than 60 years, after pooling the three cohorts, to assess confirmed psoriasis patients and the prevalence of T2D.

  1. A drug to treat attention-deficit/hyperactivity disorder (ADHD) in adults was studied in this experiment. This is an experimental study that used placebo-controlled design. I know it is an experimental study because they manipulated an aspect of the experiment (if the drug was administered or not). They used at total of 31 outpatients sites across North America as their randomized patient base. The patients used were ones who met the DSM-IV criteria for ADHD after undergoing an interview by two researchers. The patients then were normalized with a comorbid psychiatric diagnosis by clinical interviewers with the criteria by the Structured Clinical Interview for DSM-IV. The approved patients were all given an initial 1-week medication washout and evaluation period, followed by a 2-week placebo lead-in phase. This was a modified double blind (the efficacy raters were blind to the protocol, but the others at the investigative sites were not). At this point, the patients who maintained the initial criteria were used and were randomized to receive either atomoxetine or a placebo for a ten-week period. The patients were monitored biweekly and adherence to the procedure was monitored by pill counts and history.
    Statistical analysis was done based on the investigator-rated CAARS scores for the initial tests vs. the self-reported CAARS score from the patients. These were presented at t-scores and were then allowed for comparison.
    For this study, the CAARS scores are the response variable and the medication is the experimental variable.


  2. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3422463/
    This article deals with sGC, or prenatal synthetic glucocorticoids, and what the effects of pregnancy treatments with them are on methylation of DNA in the fetus and in a subsequent generation. This is an experimental study as they are actively manipulating the variables and injecting the guinea pigs with the sGCs.
    The methodology was very thorough and seemed to be appropriate for discovering the answers to the research question. For instance, they took samples at multiple intervals like during the pregnancy and from the subsequent generations. In addition, for all of these groups, they took samples at two separate time periods, one before the natural increase and one after the natural increase. Similarly, they took samples from the liver, right kidney, right adrenal, and placenta in order to compare results from different areas of the body. The DNA and RNA were extracted and examined with two methods specific to them. The choice of methodology covered all aspects and protected against a variable appearing that was not thought of prior to the experiment.

  3. This study examined trends in sociodemographic characteristics of mothers who had a preterm birth in the city of Valparaiso, Chile. The research was population-based and longitudinal, meaning it occurred over a long period of time (1991-2008). Using the National Live Birth Database of Chile, the researchers obtained data collected by midwives and doctors during the time of birth. This information included age of the mother, marital status, weeks of gestation of the baby, and weight of the baby. The study used varying data extrapolations and statistical analyses to form relationships and draw conclusions (tested for statistical significance using p scores, odds ratios, etc). Since this experiment did not manipulate any variables and instead, observed existing data for trends, it is considered an observational study rather than an experimental study.


  4. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4099224/

    This study examined the effect of chromium picolinate supplement on pancrease and macroangiopathy of Type II diabetic mice. This study qualifies as an experimental study because the mice used in the experiment were genetically engineer and not represents the whole population of mice. ( These mice were purchased). Furthurmore, the independent variable or concentrations and composition of the supplement in the diet were altered, in order to see these variabilities affect mice. The effects of the independent variables were measured based on the changing in level of blood glucose of the mice ( dependent variable). Additionally, the level of insulin was examined in addition to level of blood glucose of mice to draw a conclusion.

    • Considering this is an experimental study, I really like the fact that they have a reliable sample size ( n=50), and the bias is kept at the minimum by making sure that all the mice were male, be around the same age and weight, and were reared in the same control environment to prevent infection from pathogens. Though, for the statistical analysis part, the author only mentioned using software to do the statistical test ( reject or not reject the null hypothesis).

  5. The name of the study I chose is “The effect of therapeutic vaccination for the treatment of chronic hepatitis B virus infection.” The research question was, how effective is the therapeutic HBV vaccination? This study is an experimental study because it determines the vaccination efficiency by giving the vaccine to a set of people and not giving it to another, then drawing comparisons between those given the vaccine and those who were not. The explanatory variable is the administering of the HBV therapeutic vaccine to some of the sample set and not given to the rest of the sample set. Thirty-two people took part in this study and had to pass a certain health standard (no outside diseases or health problems beyond HBV allowed), resulting in 15 patients being given the vaccine and 17 not given it. The time period spanned 6 months, with the vaccine being given three times and after a year had passed from this study patients were looked at again. To determine how effective the vaccine was by comparing these two groups, statistics were used: Mann-Whitney U test, p-value, Fisher’s exact test, and more.


  6. http://jech.bmj.com/content/69/3/272.long

    In this study the researchers looked at the risk of influenza in different households. They took their data from the CoPanFlu-France cohort. The data was comprises of 601 households that were actively monitored for influenza-like illness. They then used structural equations to model direct and indirect relationships between infection and risk perception by looking at preventive behavior, social contacts, indoor and outdoor environment, sociodemographic factor and pre-epidemic susceptibility. This is an observational study because their data was taken from another source, and they did not influence the subjects of the study in any way.

  7. http://www.researchgate.net/publication/216813660_Latitudinal_differences_in_plant_palatability_in_Atlantic_coast_salt_marshes

    The study was title, “Latitudinal differences in plant palatability in Atlantic coast salt marshes.” This study was focusing on the how changes in latitudes effect the palatability in plants. The used site in Rhode Island and Georgia. Found common plants and consumers among these systems.Plants were flown back and fourth. Their was a lot of work put into place in order to eliminate any bias from this experiment. Plants were used that were found in both the northern and southern area of the east coast. The salt marshes were used because of consistency of plants across latitude. They also used a number of different plant species and a number of each plant. The plants were flown both north and south and replanted. They also did the experiment at different times in the year. It was a very tedious and thought out methods.

  8. This study looked at the effect on lactose digestion that combining lactase from Aspergillus oryzae and yogurt bacteria has. Thirty participants were included in the study and all experienced self-assessed lactose maldigestion. Capsules identical in appearance were administered to participants while participants consumed milk with which additional lactose had been added to. Participants received one of five capsule preparations: yogurt bacteria, 3300 FCC lactase, 9000 lactase, a combination of yogurt bacteria and 3300 FCC lactase, or a placebo capsule. The experiment lasted 5 days and two methods of measurements were used. Lactose digestion was measured via breath hydrogen tests after 6 hours of consumption. Participants also filled out questionnaires in regards to any gastrointestinal symptoms that they may be experiencing. The questionnaire was based on a scale of 0 (no symptoms) to 6 (unbearable symptoms). Data was evaluated using Friedman tests and multiple post hoc Wilcoxon comparisons. This study served as an experimental study because researchers manipulated the capsule received by participants. The administered capsules serve as the explanatory variable while the digestion response of the patient serves as the response variable.


  9. In this study, the effect of exposure to bisphosphonates on patients with osteoporosis was explored, specifically looking at the relationship between the exposure to bisphosphonates and incidence of type 2 diabetes. The response variable was whether or not the patients were diagnosed with type 2 diabetes while the explanatory variable was whether or not the patients were exposed to bisphosphonates. The study was retrospective, which usually implies that it is an observational study. Review of the methods section confirmed this thought. Data in the form of patient records was taken from the Health Improvement Network database (THIN) over the course of 15 years. No experimentation was conducted, no variables were manipulated by the researchers; therefore the study was observational. The study focused on patients who were exposed to bisphosphonates for the purposes of treating osteoporosis for at least one year versus patients who were unexposed. Several factors were considered, including age, gender, BMI, and the primary care setting while potential cofounders were also taken into account, such as hypertension, cardiovascular disease, smoking, etc. (Konstantinos, 2015). The data was analyzed for comparison between exposed and unexposed groups using x2 tests and t-tests or Mann-Whitney U test.


  10. This study examined the differences in social behaviors between different sexes of chimpanzee infants. The study focused on the Kasekela community, which is located in the center of Gombe National Park in Tanzania. The Kasekela community contains about 38 to 62 individuals, with age–sex classes composed of 6–14 adult males, 12–25 adult females, 6–14 subadult females, and 7–15 subadult males. The explanatory variable in this study was the gender of the infant chimpanzee, while the response variable of interest in this study was the number of social partners with whom an infant chimpanzee had direct interactions. The observers counted the number and type of individuals that an infant had either groomed, played with or physically contacted within the period of interest. Various factors were also taken into account that infer primate sociality, including the season, community size, total number of individuals encountered, and average and medium group size. Because this study was limited to making observations and collecting behavioral data on encounters between infants and other members of the community, this experiment is an observational study


  11. The goal of this study was to determine if immunogenicity to measles-mumps-rubella vaccine administered to infants via a disposable-syringe jet injector (DSJI) was equivalent to the MMR vaccine administered to infants via a needle and syringe (NS). This was an experimental study as the infants were randomized to receive the MMR vaccine by either DSJI or NS; the explanatory variable (vaccination via DSJI or NS) was actively controlled/manipulated by researchers. The response variable (immunogenicity) was measured as a percentage of baseline-negative infants who seroconverted for antibodies against measles, mumps, and rubella diseases. The infants in the study were normalized: each infant had already received all other routine vaccinations (minus the MMR) according to Brazil’s Basic Child Vaccination Schedule and were considered to be in good health. This study also took into account the possibility of a child having already been exposed to, and therefore expressing antibodies against, a particular virus; for instance, if a child tested positive for anti-rubella antibodies prior to vaccination, then that child was excluded from the study population for the analysis of rubella antibody (but was still eligible for inclusion in the analysis of mumps and measles antibodies).


  12. http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0102060

    This study examines the dynamics of whale sharks at the edges of their oceanic habitats. The species is a large filter feeder and it has been determined that individuals aggregate seasonally to highly productive coastal sites, performing large migrations in order to do so. When they are not present at these sites however, the whereabouts of these sharks was largely unknown. Thus, the purpose of this study was to examine how environmental factors affect the distribution of whale sharks when they are not in specific highly productive locations. Following national Portuguese laws regarding the use of vertebrates in research, the study obtained its data from sixteen years worth of observations, from 1998 to 2013, made by the pole-and-line tuna fishery in the Azores. Environmental variables such as ocean surface temperature and amount of chlorophyll were taken into consideration, using satellite derived estimates.

  13. This study investigated the effect of the introduction of antiretroviral therapy on HIV-related mortality in Malawi, Africa. These researchers utilized a demographic surveillance system in order to measure the mortality of the people in Malawi. They looked at the rate of mortality from a time before the free antiretroviral therapy was available in the area of study (August 2002) to 8 months after the clinic was opened (February 2006). In addition to this, causes of deaths were found via retrospective interviews of verbal autopsies. The researchers also looked at age, sex, cause of death, and zone of resident when analyzing trends in mortality. Since the researchers observed trends rather than manipulated variables, this was an observational study. The explanatory variable was whether the subject received antiretroviral therapy or not, whereas the response variable was HIV-related mortality. This study could not be an experimental study because it would be unethical for the researchers to manipulate and choose what individuals received antiretroviral therapy and who did not. Instead, the explanatory variable is self-selected, meaning that the subjects chose whether to receive antiretroviral therapy.


  14. http://www.ncbi.nlm.nih.gov/pubmed/25707016

    This research sought to analyze the effects of 21 metals and metalloids in the blood and urine (e.g. lead, cadmium, mercury) on birth outcomes such as the weight, length, head circumference, and gestational age of the infants. This study differs from others in the sheer number of elements tested for, and the fact that blood/urine concentrations were taken from both the father and the mother. For methods, this was a longitudinal study, in which 501 couples planning for a pregnancy were recruited from central Michigan and the Texas Gulf Coast area. Blood and urine specimens were collected by a nurse in the couples’ homes before the pregnancy and follow-up visits were conducted every 45 days. Additionally, a survey chronicling the couples’ histories was conducted, and the date of conception was taken to determine gestational age. Other information regarding the infant was collected from the hospital. This is an observational study, because the researchers did not administer certain metals to one group of couples’ and not the other,

  15. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3907774/

    This study examines the effects of interferon on B cell development. This study utilized PCR amplification, and other isolating techniques in order to develop their results. This study was measuring an experimental response, they were trying to figure out the effect of interferon on B cell development. Therefore this study can be classified as an experimental study.

Leave a Reply