Unit 2 General Principles Of Research In Nutrition Flashcards
(51 cards)
What do nutrition scientists actually do in terms of studying nutrition science
They get facts by systematically asking questions and using the scientific method to design and conduct expiriments to answer the question and test for various possible answers
What does nutrition monitoring help with
Help researchers assess nutrient status, heath indicators, and dietary intakes of the Canadian population
What are the examples of research design
Epidemiological study
Case study
Intervention study
Lab study
What is a intervention study
Uses the basic scientific method. When used in health studies it’s called RCT (randomized controlled trials) , whci hard used for testing new drugs
The researches intervene to alter peoples diet.
Have Two groups:
- intervention: make a single dietary change, have the treatment
- control group: continue with regular diet, untreated or has a placebo
Usually called a controlled trials because there a control group
After a time the two groups are compared any any diffence between the two hours is because of the single intervention
What is the scientific method
Observation and question
Hypothesis and prediction
Experiment
Results and interpretations
Hypothesis supported:
- theory developed
- new observations and questions
Hypothesis not supported:
- new observations and questions
Then back to start
What are the safeguard in intervention studies and why
To reduce possibly of changes be attributed to random chance:
- test groups have to be large enough to reduce effect of random chance influence (30 or more subjects per group)
- random placement of people into the groups to prevent bias
- two groups are compared at start of study to make sure they a similar in factors relevant to the study. They need to match in sex age and ethnicity: ex. If effect of diet on weight loss is tested, members of the two groups need to be similar weight.
What is the placebo effect
How is it overcome
Can be found in intervention studies
If ill people are given an inert substance (placebo) and are told its actually medicine, they will report they feel better
Overcome by giving the control group in the intervention study a treatment that very closely resembles the active/actual treatment
Ex. If testing effect of sugar on energy intake , The treatment group is given drink with actual sugar , control is given drink that taste and looks similar but has synthetic sweetener
What is the double blind part of intervention studies and why is it used
Used to prevent bias of investigators
Both the patient and investigator are blind/ don’t know if the subjects have the placebo or the active treatment
- if the investigator expects that the treatment will reduce pain in the patient they’ll “observe” a reduction in pain even if treatment is ineffective
- then if the patient sees this bias, as a result the patient could “improve” with the active treatment not because it was more effective but become the investigator said it was
- in the end it’s like the active treatment got another type of placebo effect
Double blind studies are the
Gold standard in drug trials or when nutritional supplements are given
When can the blinding of subjects not happen
What happens instead
In experiments that involve major diet changes blinding int always possible
But the subjects can still be assessed on a blinded fashion:
- ex. When testing effect of low salt diet on blood pressure the patient knows their diet but the investigator recording the blood pressure won’t because the blood pressure measurement can be subjective
- in case control studies if a patients diet is assesssed the investigators shouldn’t be aware of the patients diagnosis
Some Intervention studies are done without a ____
Without a control group, termed uncontrolled
Errors:
- These are less reliables than controlled studies because you can’t be sure if subjects would have responded in the same way if not given treatment
- many conditions clear up without treatment
- assessment of subjects: if BP measured in successive days, it’s goes down. If under stress due to being in doctor office it goes up then goes down as they become less stressed. Uncontrolled trials would show that anything lowers the BP
What is good about intervention studies
Can be carried out on healthy subjects to prevent diseases, or on sick ppl to test tratment (clinical study/clinical trial)
Great in showing there there is a cause and effect relation ship between a nutrient variable and disease
What is anecdotal evidence
Also called case study
Collecting ancentdoctal evidence is the simplest and least reliable way to study role of diet in disease
Investigator makes inferences of diesease from the previous experiences of one or more individuals
can go wrong:
- women who’s smokes has healthy kid, women who doesn’t smoke has kid with Down syndrome
- observing these isolated cases would suggest that smoking prevents Down syndrome , not good so the same observation should be repeated many times
Can go well:
- many ppl in US reported headaches with specific symptoms after eating in Chinese restaurants
- the cause was mono sodium glutamate (MSG) food additive
- so anectdotal evidence was useful because many similar reports were observed
Advantages and disadvantages or anecdotal evidence
Pros:
- like the MSG example, completely new and previously unsuspected relationships can be found
- side effect from new drugs can be uncovered when many reports based on individual patients are collected and a pattern emerges
Cons:
- inherent unreliability because all types of errors can occur
- error of chance: many worker exposed a a chemical get the same type of cancer, this could be chance
- error of bias in the observation: due to the power of suggestion, people are unreliable in describing their symptoms. especially symptoms of headaches and caught pains
Overall how is anecdotal evidence used and accepted
Only accepted with lots of caution because many errors.
Provides suggestions for further study but is rarely the basis for firm conclusions
Advertisers use anecdotal evidence to promote products. This is called testimonials where they use personal stories to promote wieght loss diets, foods, or supplements.
Need to be skeptics of ads that rely on anecdotal evidence
What are epidemiological studies
The study of the incidence of distribution of diseases and studying their control and prevention in whole populations
Reveals correlation: ex. Country food has more nutrient x and these people suffer less illnesses y (inverse/negative correlation)
Four types of epidemiological studies:
- population
- historical
- case control
- cohort
What is a population epidemiological study
Give example
Compare disease patterns with various factors like food consumption on a countrywide basis
Ex.
- want to know relationship between potato consumption and pancreatic cancer
- taking data on the quantity of potatoes eaten in countries A and B and compared it with the occurrence of pancreatic cancer
- observed that people in country A eat twice as much potatoes than B, and that A have double incidence of pancreatic cancer
- on the surface this suggests that potatoes cause pancreatic cancer
What are the two problems with the population study with potatoes and cancer in epidemiological studies
Errors in the data:
- closer examination of data shows that A are rich and waste half their potatoes and B are poor and can diagnose half of their cases of pancreatic cancer
- this means both actually eat similar amount of potatoes and similar frequency of pancreas cancer
Confounding variables:
- closer examination of data shows A have more obesity and B obesity is uncommon
- the confounding variable (obesity) might be the factor causing cancer (not the potatoes)
- confouding variables occur with the factor under study and mask the real cause of the disease . Very common for this to happen in diet studies since many things contribute to diet
- more common limitation is the confounding variable
Overall what is useful about population studies
Researchers can show association:
- as one factor changes (more potatoes) disease variable changes (more cancer)
- BUT being accosiated doesn’t mean one thing caused the other. Only intervention trial can tell cause and effect
Even if there are errors it can give valuable finding :
- benefit of Mediterranean diet discovered when found that counties that border the Mediterranean have lower rate of heart disease
What is a historical epidemiological study
Give example
Taking the history of a disease to reveal clues of the cause of the disease
Useful when there are accurate records of number of people who died from or have a specific health problem
Ex. Relationship between cigarette smoking and lung cancer:
- lung cancer used to be rare, but became more common among men
- increasing trend of lung cancer in men also matches increase in smoking
- increase in lung cancer in women started after the men
- consistent with the fact that women started smoking after men
Pros and cons of historical epidemiological study
Historical studies are best compared with population studies
Pros:
- historical studies are based on large populations and involve large changes in disease incidence
- can use as a valuable clue to support a theory
Cons:
- disease stats taken 60 years ago can be unreliable
- confouding variables can cause fake relationships, meaning there can be guilt by association:
- ex. Disease that increased in incidence in the 20th century will be correlated with any lifestyle factor that also changed
- so it gives accosition between factors but not definitively conclusive
What are case control and cohort studies in Epidemiological studies generally
In both, looking at individual in a single population instead of making comparisons of different populations
These studies try to explain why particular individuals in a population suffer from certain diseases based on their exposure to factors thought to be related to the disease
What are cohort studies specifically
Also called prospective studies
Large number of healthy subjects are asked questions to see each persons exposure to the factors of the study :
- so with diet studies they determine the exposure by either 24hr food recall or by a food frequency questionnaire
- investigators keep track of the subjects over the years until enough of them developed the disease of interest ( ie. The study variable of interest are not manipulated, but actually monitored overtime)
- then the investigators compare the ones that got the disease to those that didn’t, taking into account exposure to the disease causing factors
Give an example of a cohort study
Ex 1. Nurse’s health study in the US where 80,000 healthy nurses did survey on health status and lifestyle with a food frequency questionnaire
They were surveyed a gain after 10 year with similar questions and got back a bunch of data about the relation of cancer and heart disease with lifestyle
Ex 2. Relationship between milk product consumption and growth in a cohort of children