2. Use this extract taken from the article, “Combination of Garlic and Lemon juice prevent blood sugar of diabetic patients,” (appeared in Diabetes Research on October 31, 2016) to answer the questions that follow:

Diabetes is considered as one of the major public health problems worldwide. It is a chronic disorder in which a person has high blood sugar because the body does not produce enough insulin. Diabetes is the major cause of mortality and morbidity in the world. A recent study by researchers at the US Diabetes Research Centre and Oklahoma University School of Public Health investigated the effects of a mixture of garlic and lemon juice on diabetic patients in the US.

A double-blinded, randomized, placebo-controlled trial was conducted with 360 diabetic patients. Randomization was stratified by gender. Patients were randomly assigned to one of four groups. Each group consisted of 45 men and 45 women. The first group was assigned to receive the mixture of garlic and lemon juice daily, the second group was given 20 grams fresh garlic daily, the third group was given 1 tablespoon lemon juice daily, and the last group was not given any intervention. After fifteen weeks of intervention, it was found that the HbA1C (three-month average sugar level) of the combined garlic and lemon juice group was significantly lower compared with the other groups.

The researchers also found a greater reduction in BMI, systolic and diastolic blood pressure in the mixed group compared with garlic, lemon juice and control groups.

(a) Is this an experimental or observational study? In less than 50 words clearly explain your choice based on the extract given above.

(b)  For the above study identify, if appropriate, the

i) response variable(s).

ii) factor and its levels.

iii) sample size.

(c) Are the four principles of experimental design used in this study? Explain, in the context of the study?

(d) Explain explicitly what a confounding variable is. Identify one plausible confounding variable in this study and explain why it is a confounding variable.

Answers

1a. The appropriate model to represent the variable of interest in the given case would be binomial distribution. The relevant parameters associated with the model are probability of success (p), number of trials (n) and number of successes (x) that are desired.

 b. The conditions of binomial distribution are satisfied in the given case as indicated below.

• The given study consists of 20 trials being conducted.
• There are only two potential option for the selected sample i.e. the given person is obese or not.
• Also, the success probability i.e. probability of the person being obese is independent for each trial and is not impacted by the previous outcome.
• Independence of the trials is also maintained as the outcome on one of the trials has no impact on the outcome for the other trial.

c. Mean = n*p = 20*0.7 = 14 people

Standard Deviation = [np(1-p)]^0.5 = (14*(1-0.7))^0.5 = 2.05 people

d. The required probability = 1- P (X≤14) = 1- BINOMDIST(14,20,0.7, TRUE) = 1-0.5836 = 0.4164

e. Requisite probability = 1- P (X≤74) = 1- BINOMDIST(74,100,0.7, TRUE) = 1-0.8369 = 0.1631

 The necessary conditions to be tested are as follows. 

• The success probability i.e. probability of the person being obese is independent for each trial and is not impacted by the previous outcome.
• Independence of the trials is bein maintained as the outcome on one of the trials should have no impact on the outcome for the other trial.

2a. The given study is experimental study as in an observational study treatments are not assigned to the subjects and it is not under the control of the researcher. However, the study indicated here is a controlled trial and treatments are assigned to the various participants by the researcher.

b. (i) Response or dependent variables in the given study are HbA1C (three-month average sugar level), systolic and diastolic blood pressure and also BMI.

(ii) The various factors along with the corresponding levels are as stated below.

• Mixture of garlic and lemon juice (Frequency- Daily)
• Fresh Garlic (Frequency- Daily, Quantity – 20 grams)
• Lemon Juice (Frequency- Daily, Quantity – 1 tablespoon)

(iii) Each group has 90 members i.e. 45 males and 45 females. Since there are four groups in total, hence overall sample size = 90*4 = 360

c. The applicability of the four factors of experimental design for the given case is explored below.

• Controlling – This given study does not do well on this count as there is no effort undertaken to prevent the influence of other factors such as food undertaken before and after the intervention that may significantly impact the outcome of the study.
• Randomization – The given study takes into consideration randomization as the test subjects are randomly allocated to the four groups which helps in keeping under check the influence of difference in dietary habits.
• Replication – The given study takes replication into consideration as the sample selected for each group is sufficiently large and accounts for both the genders, hence ensuring higher chances of the study being replicated with comparable results.
• Blocking – The given study takes this into account which is apparent from the fact that in each group there are equal males and females. This is indicative of the fact that the researcher expects that gender would be a significant parameter influencing the outcome and hence has included equal representation of the two genders.

d. Confounding variable may be defined as a variable which tends to act as the mediator and influences the relationship between the independent and dependent variable. One of the confounding variables in this case could be the other food consumed by the participants during the given period. It is likely that there are some individuals who tend to consumer healthy food would more likely show improvement in the various dependent variables of interest in the given case. Hence, it is possible that the underlying relation may be dependent on the underlying food habits.