Association Between Dietary Flavonoids And Age-related Macular Degeneration (AMD)

Independent variable and dependent variable of the study

(i) Identify the study design used in this study and briefly explain (in your own words) how this specific study design was used to carry out this piece of research. 
(ii) Identify the independent and dependent variable(s) in this study. You must be specific with your answers to get full marks.
(iii) Use paraphrasing to explain what the objective of this study was in your own words. For this question you are expected to demonstrate a clear understanding of the objective/aim of the study.

Q. 2.
The methods section of the study states:
“Dietary data were collected with the use of a 145-item self-administered food-frequency questionnaire (FFQ)…Participants used a 9-category frequency scale to indicate the usual frequency of consumption of individual food items during the past year,” (Gopinath et al. 2018, p. 2)
(iv) Describe two types of bias which could be at play given the above situation. (NOTE: You have been introduced to two types of bias in this unit, however, you may have to use credible sources such as reputable websites or textbooks to investigate other types of bias to inform your answers to this question).
(v) Based on what we have learned about scales of measurement in this unit, name the scale of measurement used by implementing a ‘9-category frequency scale’ and briefly justify your answer. 

Q. 3.
(i) “In exploratory analyses, we assessed associations with the prevalence and 15-y incidence of AMD, which were the primary endpoints,” (Gopinath et al. 2018, p.3).
Using credible sources such as reputable websites or textbooks investigate the meaning of the terms ‘incidence’ and ‘prevalence’ in the context of health research and explain what they mean in relation to this study regarding AMD. 
(ii) In the statistical analysis section of the paper the authors describe how they controlled for a number of confounding factors. Name three of the factors the factors they controlled for in this study.

(iii) Select one of the factors you identified for

Q. 3 (ii) above and using the triangle model (which is explained in the unit content) describe how the factor you selected meets the three criteria for a potential confounding factor with respect to this AMD related study. You
should provide a simple triangle model which is appropriately labelled and use this as an aid to explain the how the chosen factor meets the three criteria to be a potential confounding factor.

‘Table 2: Associations between flavonoid intake and the prevalence of AMD in the Blue Mountains Eye Study’ (Gopinath et al. 2018, p.4) shows the association between levels of intake of flavonoids and the prevalence of AMD. Participants in the highest quartile had the highest intake of total flavonoids compared with those in the lowest quartile who had the lowest intake of flavonoids. Odds ratios and confidence intervals are the key statistics used to present the data in this table. The first quartile for each flavonoid group is used as the reference or baseline data and this has an odds ratio of 1, this represents a standard or baseline level of risk of AMD. Please study the table carefully
before answering the following questions.

Main objective of the study

Q. 4.
(i) Based on the data regarding ‘All flavonoids’ in the ‘Any AMD’ column, identify the odds ratios for each of the second, third and fourth quartiles. (3 marks)
(ii) Based on what you have learned about odds ratios in HBS108, clearly explain in your own words what these three odds ratio values [from Q. 4 (i)] tell us about the prevalence of AMD and ‘All flavonoids’ consumption in the study participants.
(iii) Based on the confidence intervals (CIs) attached to each of the three odds ratios respectively [from Q. 4 (i)] are these particular study findings considered statistically significant Use what you have learned about statistical significance and CIs in this unit to explain how you interpreted the CIs. The relevant CIs must be stated in your response to this question.
(iv) Write a statement explaining what statistical significance means in the context of AMD prevalence and ‘All flavonoids’ consumption in this study. 

Q. 5.
(i) Based on the data regarding ‘Total flavanones’ in the ‘Any AMD’ column identify the odds ratio which relates to the fourth quartile and explain what this odds ratio tells us in the context of these study findings. (3 marks)
(ii) Based on the data regarding ‘Total flavanones’ in the ‘Any AMD’ column identify the CI associated with the odds ratio of the fourth quartile and explain what this CI tells us in the context of these study findings.

Q. 6.
“Our study suggests that the consumption of oranges (a key contributor to total flavanones) is inversely and independently associated with both the prevalence and incidence of late AMD,”(Gopinath et al. 2018, p.5). Explain in detail what is meant by the phrase “inversely associated”within the context of this specific statement about oranges and AMD. 

Q. 7.
The following article was written in light of the findings of the study by Gopinath et al. (2018):Logan, T 2018, An orange a day keeps the macular degeneration away, early research finds, media release, 18 July, ABC News South West WA, 
Discuss in detail the accuracy of the title of the above media release in relation to the findings of the research study by Gopinath et al. (2018). NOTE: You should use this question as an opportunity to demonstrate your understanding of the skills and concepts taught in HBS108. It is essential that you support your answers by backing them up with specific pieces of evidence from the results and discussion sections of the research study. At least three and distinct well explained reasons must be provided as a rationale for this answer. 

Q. 8.
Apply the Bradford-Hill test to identify if causation could be a possibility based on this study between dietary flavonoids and AMD. For this question you should provide a very detailed response to each of the criteria/questions of the Bradford-Hill test using specific information from the study to support each of your responses.

Q. 9.
Using your responses to questions 7 and 8 to inform detailed answers for the following set of questions:
(i) Propose a suitable study design for future research into dietary flavonoids and AMD.
(ii) Explain, with specific details, how your proposed study [from Q. 9 (i)] could make a valuable contribution or advancement to the existing body of evidence on dietary flavonoids and AMD. Your answer should be backed up with reference to your responses to Q. 7 & 8.
(iii) Explain any practical or ethical challenges researchers would have to consider when carrying out the research study you suggested for Q. 9 (i).

Independent variable and dependent variable of the study

The study design was population-based cohort study with cross-sectional analysis. In the research, Gopinath et al. (2018) recruited 2856 adults aged greater than or equals to 49 years at baseline (for prevalence analysis) and 2037 adults for 15 year follow up (for incidence analysis). The prevalence and incidence analysis helped to ascertain the association between age-related macular degeneration and dietary intake of flavonoids.

Answer 1 (ii)

Independent variable of this study is flavonoids. Depended variable is age-related macular degeneration (AMD). This is because, the independent variable (flavonoids) is changed or controlled by the researcher and the effects of changing the independent variable is reflected on the dependent variable (AMD), the one which is measured (Hayes & Preacher 2014, p. 451).

Answer 1 (iii)

The main objective of the study is to access the independent association between the prevalence and 15-year incidence of AMD with the dietary intake of total flavonoids and common flavonoid (Gopinath et al. 2018, p. 1).

Answer 2 (iv)

First bias is confounding bias

Confounding bias because, the participants selected are either predisposes to AMD or consumes certain amount of flavonoids helpful in reducing the threat of AMD. So in this case, age and the previously known consumption of flavonoids is a confounder for the effect of total flavonoids on the development of AMD (Gopinath et al. 2018, p. 3).

Second bias is information bias

Information bias because, the participants were given self-administered food-frequency questionnaire (FFQ). This might increase the chance is systematic errors in measurement or misclassification of the exposure (Gouw et al. 2013, p. 232).

Answer 2 (v)

The scale of measurement used by implementing "9 category frequency scale" is interval scale. This is numeric scale used for statistical analysis. Since the data obtained are quantitative like first quartile, second quartile, third quartile and fourth quartile, the scale is quantitative and hence interval scale.

Answer 3 (i)

Incidence means rate of occurrence. That is number of individuals who develop AMD during older adult stage (Parisi et al. 2013, pp. 378).

Prevalence means total number of individuals in the population who have AMD at a specific period of time and is expressed in percentage of population (Parisi et al. 2013, pp. 378).

Answer 3 (ii)

Three factors controlled in the study include age, sex and covariates which is found to be associated with the incidence of AMD (smoking, consumption of fish and intake of zeaxanthin and lutein) (Gopinath et al. 2018, p. 3).

Main objective of the study

Answer 3 (iii)

Answer 4 (i)

0.63, 0.62 and 0.52 respectively (Gopinath et al. 2018, p. 4)

Answer (ii)

With the increase in the All flavonoids consumption, the odds ratio is decreasing therefore there is inversely proportional relationship. Thus the outcome is 48% less likely to develop AMD in the fourth quartile (Gopinath et al. 2018, p. 4).

Answer 4 (iii)

Odds ratio are used for measuring the association between a outcome and the preceding exposure (TchetgenTchetgen 2013, p.4570). Thus, reduction in the OR from the third to the fourth quartile indicates a less likely chance of AMD with intake of flavonols. Statistical significance refers to the likelihood of a relationship between variables being caused due to an event, in place of chance. Conversely, confidence interval refers to the range of values where the true value is present (Altman et al. 2013, p.552). Hence, 95% CI indicates 0.95 probability of comprising of the population mean .

Answer 4 (iv)

Statistical significance mean that upon increasing the intake of all flavonoids, the likelihood of onset of late AMD is low among people (Gopinath et al. 2018, p.1).

5 i) OR for total flavonones in the any AMD column, corresponding to the fourth quartile is 0.29. This indicates that the group shows a reduced likelihood of experiencing the particular event (AMD)(TchetgenTchetgen 2013, p.4569).

Answer 5 (ii)

The CI for the fourth quartile corresponding to total falvonones for any AMD is 95%. A 95% confidence interval comprises of a series of values that confirm 95% consistency of containing the actual mean of this population(Gopinath et al. 2018, p.4). 

Answer 6 (i)

An inverse association is also referred to as a negative correlation and the term is used for explaining a contrary relationship between a dependent and an independent variable that are moving in opposite directions (Silbert et al. 2013, p.819). Thus, with an increase in the consumption of oranges that contain total flavonones, there was a reduction in the prevalence and/or incidence of AMD.

Answer (7)

The title of the newspaper article was accurate in stating that consumption of an orange every day plays an effective role in lowering the risks of onset of macular degeneration. This can be explained by the fact that in the research, 1 increase in standard deviation in the total amount of overall flavonoid was significantly related with lower risks of AMD (multivariable-adjusted OR: 0.76; 95% CI: 0.58, 0.99)(Gopinath et al. 2018, p.3). Thus, the news article was correct in establishing flavonoids present in oranges as the key component (Logan 2018).

Biases in the study

The role of flavonoids in reducing oxidative stress and inflammation in the body was accurately presented owing to its impacts on enhanced vascular health(Gopinath et al. 2018, p.2). The article also focused on the drawbacks of the observational study.

This was explained in the research paper by the statement that the observational study failed to distinguish the impact of residual confounding from factors that were unmeasured such as, inflammatory markers(Gopinath et al. 2018, p.6).

Answer (8)

The Bradford-Hill criteria is given below (Fedak et al. 2015, p.14):

• Strength- Moderate
• Consistency- No
• Specificity- Yes
• Temporality- Yes, effect occurred after cause(Gopinath et al. 2018)
• Biological gradient- Inverse proportion
• Plausibility- Limited
• Coherence- Lack of coherence
• Experiment- Yes
• Analogy- Yes

Answer 9 (i)

Prospective cohort and/or epidemiological studies need to be conducted

Answer 9 (ii)

Prospective cohort studies will follow similar individuals over a time frame and determine the changes in the dependent variable (AMD outcome) with respect to flavonoid consumption. Epidemiological investigation will establish a link between risk factor and the incidence of disease that is being investigated (AMD)

Answer 9 (iii)

Some of the challenges that might be encountered are as follows:

• Following large number of sample for long duration
• Time consuming and expensive
• Differential loss might lead to bias
• Constraints due to sudden changes in nutritional status and health 

References

Altman, D., Machin, D., Bryant, T. & Gardner, M. eds., 2013, Statistics with confidence: confidence intervals and statistical guidelines, John Wiley & Sons.

Fedak, K.M., Bernal, A., Capshaw, Z.A. & Gross, S., 2015,‘Applying the Bradford Hill criteria in the 21st century: how data integration has changed causal inference in molecular epidemiology’. Emerging themes in epidemiology, vol.12, no.1, p.14,

Gopinath, B., Liew, G., Kifley, A., Flood, V.M., Joachim, N., Lewis, J.R., Hodgson, J.M. & Mitchell, P., 2018,‘Dietary flavonoids and the prevalence and 15-y incidence of age-related macular degeneration’, The American journal of clinical nutrition, vol.108, no.2, pp.381-387,

Gouw, S.C., Van Der Bom, J.G., Ljung, R., Escuriola, C., Cid, A.R., Claeyssens-Donadel, S., Van Geet, C., Kenet, G., Mäkipernaa, A., Molinari, A.C. & Muntean, W., 2013, ‘Factor VIII products and inhibitor development in severe hemophilia A’, New England Journal of Medicine, vol. 368, no. 3, pp.231-239. DOI: 10.1056/NEJMoa1208024

Hayes, A.F. & Preacher, K.J., 2014, ‘Statistical mediation analysis with a multicategorical independent variable’. British Journal of Mathematical and Statistical Psychology, vol. 67, no. 3, pp.451-470.

Logan, T 2018, ‘An orange a day keeps the macular degeneration away, early research finds’, ABC News, 18 July, 

Parisi, R., Symmons, D.P., Griffiths, C.E. & Ashcroft, D.M., 2013, ‘Global epidemiology of psoriasis: a systematic review of incidence and prevalence’, Journal of Investigative Dermatology, vol. 133, no. 2, pp.377-385. 

Silbert, B.I., Pevcic, D.D., Patterson, H.I., Windnagel, K.A. & Thickbroom, G.W., 2013,‘Inverse correlation between resting motor threshold and corticomotor excitability after static magnetic stimulation of human motor cortex’, Brain stimulation, vol.6, no.5, pp.817-820, 

TchetgenTchetgen, E.J., 2013,‘Inverse odds ratio?weighted estimation for causal mediation analysis’. Statistics in medicine, vol.32, no.26, pp.4567-4580,