Statistics Report on Randomized Control Trial
- 20th Sep, 2022
- 15:46 PM
Title: Randomized control trial
There is a claim by a newly organised group of people of a company that they can conduct a program that is beneficial for weight loss. This needs to be statistically tested to know if there is a fact in the claim of the group.
Study question: Test whether new and improvised treatment claimed by the company has a significant effect on weight loss than ordinary treatment.
The study is conducted in the most populated city in the US with an estimated population size of 3 million
Intervention: Patients are advised to follow treatment protocol for six months, each day the patients of each group will report their weight and food they have consumed and are kept on physical examination such as systolic and diastolic blood pressure, body measurements, height and weight measurements, body comparison measures and discuss their progress. All the treatment assessments must be taken by an independent evaluator who will not have knowledge of treatment conditions.
Control Groups: One group will be on the ordinary program and another group will be on a new treatment program.
We measure BMI as continuous measures from all the patients and also note blood pressure, blood glucose level, cholesterol level, waist/hip circumference and diet quality sources.
Base design: We design the test so that both groups are assessed in parallel for separate treatment plans. The assessment is carried out by an independent evaluator who does not know about treatment conditions. Participants' age must be taken between 20 and30, participants will be randomized in a 1:1 ratio based on the SAS program. We first divide all the randomized patients into two groups based on whether BMI is in (between 28 and 32) or (between 32 and 40). Each group is then analysed in parallel for separate treatment plans and significance is analyzed using inferential statistics. We use stratified sampling for the randomization process to avoid imbalance in groups.
Sample size calculations:
(1) We will target to detect if the standardised mean difference in BMI of the two groups is greater than 8.
(2) We will choose the effect size of control groups between 0.6 which indicates a relatively higher effect size.
(3) We expect a standard deviation of 8 between the two groups since this should also replicate the population standard deviation.
(4) Sample size must be considered with a safety factor of 10% so that even if 10% dropped out after selecting the participants, the desired size will be maintained even after drop out.
(5) Calculation of sample size requires factors of standard deviation, standardized mean difference and effect size. After acquiring this knowledge we can calculate size from this.
(6) SAS code for sample size calculation:
• proc power;
• twosamplemeans test = diff
• meandiff = 2
• stddev = 8
• power = 0.80
• npergroup = . ;
Data Analysis Plan:
We collect the samples of two treatment groups and then organise all the data after contributing to missing values. Conduct statistical tests using tools like excel and conduct inferential statistics to find the group that has significantly higher weight loss as compared to the new experimental plan. In the case of variables like BMI which are continuous variables, we use ANOVA whereas in the case of categorical variables we use the chi-square test. In the statement of the problem, the null hypothesis will be there is no difference in weight loss in the case of both treatments and the alternate hypothesis will be that the new treatment plan has a considerable effect. We will consider the test at a 5% level of significance. The variance of the population in each group is assumed to be the same and also follows a normal distribution
Deborah et.al (2003) said that health improvement plans on the Internet seem promising for transient weight reduction yet have not been read for weight reduction in people in danger of type 2 diabetes; along these lines, the more extended term adequacy is obscure. To look at the impacts of an Internet health improvement plan alone versus with the expansion of social advising by means of email accommodated 1 year to people in danger of type 2 diabetes. Participants were randomized to a fundamental Internet or to an Internet in addition to conducting the e-directing project. The two gatherings got 1 up close and personal guiding meeting and a similar centre Internet programs and were told to submit week after week loads. Members in e-advising submitted calorie and exercise data and got week-by-week email social guidance and input from a counsellor. The Internet offers chances to create conduct change intercessions limiting up close and personal communication. It has been utilized for diabetes instruction and self-management,6,7 and we have utilized the Internet to convey a conduct health improvement plan with good present moment results.8 However, the adequacy of Internet-based get-healthy plans and explicitly email guidance has not been utilized in a populace in danger of diabetes nor assessed for a year-long weight reduction intercession.
Mary et.al 2015 stated that accomplishing upkeep of weight reduction is significant to battling corpulence. In any case, most people will in general recover weight. Information from fruitful maintainers shows that they stay cautious and continually apply methods to restrict the course of recapturing. Then again, ebb and flow propel in heftiness research show that the decreased fat state is a condition of modified physiology as far as vitality balance. This audit depicts the physiological adjustments happening after weight reduction that incline to recovery. In particular, changes with respect to body piece, hormonal foundation, vitality consumption and control of food admission are examined. In addition, metabolites can go about as recapture indicators and dietary procedures to restrict recovery are introduced