The researchers conducted a single-site parallel-group trial for weight loss where participants were randomized into either the group receiving reduced fat or reduced carbohydrates diets for a period of twelve months. The study’s aim was to determine if a group comprising of 3 SNP genotype design or the baseline variations in release of insulin or the two were predisposing factors to differences in success for twelve months while either on a low fat or carbohydrate diet (Gardner et al., 2018). The study had two hypotheses. The first hypothesis stated that there existed a significant relationship between diet and the pattern of genotype in losing weight. The other hypothesis stated that there existed a significant relationship between diets and the release of insulin in losing weight. The two main hypotheses were addressing twelve months of changes in weight through diet, genotypes and diet, and INS-30 identified at baseline in relation to diet (Gardner et al., 2018).
Methods
Sample
Participants were recruited through advertisements placed on media platforms. Also, emails from past recruitments for other nutrition-related research implemented by the same institution were used for recruitment purposes. A total of 609 individuals participated in the study.
Delegate your assignment to our experts and they will do the rest.
Inclusion and Exclusion Criteria
Males and premenopausal females between the age of eighteen and fifty years with a BMI between 28 and 40 were recruited for the study. However, those found to have uncontrolled high blood pressure, metabolic diseases such as diabetes, pregnant, and lactating were excluded from the study. In addition, individuals who were on medications that have an impact on body weight or the energy consumed, such as antihypertensive and hypoglycemic drugs, were excluded from the study. Other types of drugs not identified were allowed if individuals were in a stable condition when on the medication for not less than three months before the baseline data was collected (Gardner et al., 2018).
Randomization
Participants for the study were randomized into either a reduced-fat or reduced-carbohydrate diets groups. Randomization was conducted by allocating sequence generated by a computer in block sizes of eight. The statistician involved in randomization was not part of intervention implementation or collection of data. All the participants were not aware of the groups they were assigned until the required baseline data was collected, and the first session for the intervention attended.
Data Collection and Measurements
Data was gathered at baseline and at three, six, and twelve months for all the participants. Data on dietary intake was collected by using recall interviews. Energy expenditure data were collected using Stanford Seven-Day Physical Activity Recall questionnaire (Gardner et al., 2018). Participants’ weight was determined using a digital scale while Affymetrix UK Biobank Axiom microarray assessed SNPs (Gardner et al., 2018). Each secondary outcome was examined using standard assessment methods. Adult Treatment Panel III guidelines were applied in determining metabolic diseases.
Analysis
Statistics were simulations based. A two-sided Wald test was conducted at a significance level of .05. Generalized, linear mixed-effect frameworks were used to address hypotheses (Gardner et al., 2018). Intent to treat concepts were applied. Satterthwaite approximation was applied in on the Wald tests. The P values derived were all descriptive. Statistical analyses were done using R version 3.4.0. For the mixed-effects models applied in analysis, the researchers used the lme430 package. However, for mixed-effects models that were used in testing the hypotheses, the lmerTest29 package was applied.
Findings of the Study
There was no statistical significance for the relationship between diet and genotype patterns throughout the twelve-month period. Also, a significant difference in loss of weight for a period of twelve months was not identified. A significant interaction between 3 SNP genotypes and diet was not observed. The same case was identified for diet and insulin release at baseline. For secondary outcomes, a significant difference between diet categories and blood lipid levels was identified. However, although there were improvements in diet groups, no significant differences were observed among groups in their BMI, body fat composition, and waist circumference. Changes in lipids were favorable to the reduced fat-diet participants because of improvement in low-density lipoproteins and for the reduced carbohydrate group because of improvements in HDL and triglycerides (Gardner et al., 2018). Therefore, the study results indicated that there was a significant variation in weight changes between the low fats and carbohydrates diets groups. In addition, different genotype patterns were not an influence on the dietary impact on losing weight. The two identified hypotheses were not supported by the study’s findings because there were no significant differences in the main study variables.
Potential Threats to Internal and External Validity
Internal validity refers to the level which one can be confident of the fact that the cause and effect interaction identified is not explainable through other different factors ( Clark-Carter, 2018 ). Internal validity ensures that the conclusions made are both trustworthy and credible. Based on the fact that the study involved multiple groups, the potential sources of threat include selection bias, instrumentation, and lack of some important data for some participants. Selection bias indicates that the groups involved are not comparable to the beginning of a research study ( Clark-Carter, 2018 ). The study was conducted in an area that is characterized by individuals of a higher socioeconomic status and education levels. As a result of the differences, improvements in the various groups might have been as a result of other factors besides the intervention ( Leavy, 2017 ). Also, other factors were not put into consideration as well as the role played by insulin in the intervention. Some data for 78 participants were missing, and this may have had an effect on validity.
The Stanford 7-day Physical Activity Recall instrument does not offer specific examination of energy use. Social interaction could also have been a threat to internal validity. External validity is the level which findings of a study are generalizable in other contexts or on other people ( Nicholson, 2018 ). The external threat to the study’s validity was sampling bias. Sampling bias occurs if the study sample is not representative ( Baldwin, 2018 ). The fact that the study was done in a geographical area characterized by people with higher levels of education and better income indicates that most participants were most likely from the area. Therefore, the sample was not representative, and the findings cannot be generalized on low-socioeconomic populations.
Conclusion
In order to address the threats to validity, the researchers should have conducted the study in a geographical area with a high possibility of getting a representative sample. This would have prevented sampling bias. A representative sample allows the generalization of study results because all the individuals in the large population are represented in the study. Recalibration would have helped reduce selection bias. It involves the use of algorithms in order to correct weighting factors ( Nicholson, 2018 ). In addition, specific measurement tools should have been used, and all the confounding factors controlled for the study. This would have helped in ensuring that the findings cannot be explained by other factors other than the intervention used in the study. Blinding the study participants on the aims of the study helps prevent any impact of social interaction between participants ( Dunbar-Jacob, 2012 ).
References
Baldwin, L. (2018). Internal and external validity and threats to validity. Research Concepts for the Practitioner of Educational Leadership , 31-36. https://doi.org/10.1163/9789004365155_007
Clark-Carter, D. (2018). Research designs and their internal validity. Quantitative Psychological Research , 41-61. https://doi.org/10.4324/9781315398143-4
Dunbar-Jacob, J. (2012). Minimizing threats to internal validity. Intervention Research . https://doi.org/10.1891/9780826109583.0006
Gardner, C. D., Trepanowski, J. F., Del Gobbo, L. C., Hauser, M. E., Rigdon, J., Ioannidis, J. P., Desai, M., & King, A. C. (2018). Effect of low-fat vs low-carbohydrate diet on 12-Month weight loss in overweight adults and the association with genotype pattern or insulin secretion. JAMA , 319 (7), 667. https://doi.org/10.1001/jama.2018.0245
Leavy, P. (2017). Research design: Quantitative, qualitative, mixed methods, arts-based, and community-based participatory research approaches . Guilford Publications.
Nicholson, W. K. (2018). Minimizing threats to external validity. Intervention Research and Evidence-Based Quality Improvement . https://doi.org/10.1891/9780826155719.0009
