Association of Fish Consumption with Incidence of Type 2 Diabetes: A Prospective Study

This article discusses a prospective study that investigates the association of total fish consumption with incidence of type 2 diabetes. The study clarifies that preparation method is more important to find association between fish consumption and type 2 diabetes compared to overall fish consumption. The study population included Swedish men between the age of 45-79 years residing in Orebro and Vastmanland counties of central Sweden. The study design was a prospective population based study. The main research factor of interest in the study is dietary intakes and consumption of fish is being reviewed because contamination of foods and frying of foods results in formation of advanced glycation end products, mutagenic compounds and changes in fatty acid composition. The main outcome factor is the risk of diabetes. The article also discusses the presence of residual confounding and selection bias in the study.

Running head: EVIDENCE BASED NURSING RESEARCH
Evidence based nursing research
Name of the student:
Name of the University:
Author’s note

1EVIDENCE BASED NURSING RESEARCH
Answer 1 (Research question)
The main research question is to investigate about the association of total fish
consumption with incidence of type 2 diabetes in a large population based prospective study. The
researcher considered exposure to contaminants in fish to frame the research question.
Answer 2 (Study factor and outcome factor)
The main research factor of interest in the study is dietary intakes and consumption of
fish is being reviewed because contamination of foods and frying of foods results in formation of
advanced glycation end products, mutagenic compounds and changes in fatty acid composition.
Furthermore, method of preparation might also be linked to type 2 diabetes. There is lack of
studies explaining links between fish consumption and risk of diabetes. Hence, fish consumption
is the main study factor and risk of diabetes is the main outcome factor.
Answer 3 (Study population)
The population for the study included Swedish men between the age of 45-79 years
residing in Orebro and Vastmanland counties of central Sweden.
Answer 4:
The study population is representative of the population of interest as they were
included in the study based on age distribution, educational level and prevalence of overweight.
The age distribution and location of residents may help to evaluate food consumption pattern in
the population and its link with type 2 diabetes.
Answer 5 (Study design)

2EVIDENCE BASED NURSING RESEARCH
The investigation regarding the topic was done by the use of prospective population
based study. It is a type of epidemiological study design where the main purpose is to look for
outcome during the investigation process and link it to suspected risk factors. It is done by taking
cohort of interest and observing them for a long period. As the main objective of research was to
explore about effect of fish consumption (risk factor) on type 2 diabetes (outcome), prospective
study design is the best design to gain answer to the research question. The research is designed
specifically to evaluate outcome and the research design is very relevant to the study topic.
Answer 14
In epidemiological research, strength of association is dependent on strong correlation
between the cause of disease and the occurrence of the disease relative to the hypothesized
causal agent. Cohort based studies helps in accurate interpretation of causal association between
exposure and outcome. The follow up with study participants for 15 year (from 1 January 1998
to 31st December 2012) showed that men who consumed fish had higher energy intake and they
were also slightly older and with university education. The hazard ratio was assessed with 95%
confidence interval for the association between fish consumption, fried fish consumption and
specific fish items and incidence of type 2 diabetes. The statistical analysis of the data revealed
statistically non-significant association. Fish is the main source of exposure to polychlorinated
biphenyls (PCBs) and methyl mercury (MeHg), the dietary source associated with type 2
diabetes. However, the study revealed no association of PCBs and MeHg with type 2 diabetes.
Strong association was not found for men with high fish consumption, however strong
association was found between fried fish consumption and higher incidence of type 2 diabetes in
the population group. Hence, the research clarified that preparation method is more important to

3EVIDENCE BASED NURSING RESEARCH
find association between fish consumption and type 2 diabetes compared to overall fish
consumption.
Answer 15
Residual confounding is those factors that remain even after controlling for
confounding factors in research. Although total fish consumption, method of preparation and
types of fish consumption was assessed, however the difference in exposure to PCBs and MeHG
might be residual confounding that can influence study outcome. Appropriate strategies were not
taken to consider about specific contaminants present in food.
Answer 16
The study gave the conclusion that preparation method is most important to judge
about the association between fish consumption and incidence of type 2 diabetes. This
conclusion is valid because no association has been found between fish consumption and types 2
diabetes, however strong association has been found between preparation method and risk of
diabetes. The conclusion also gives the implication regarding increasing awareness about
preparation method to control diabetes.
Selection bias:
The study participants were recruited in terms of age distribution, education ad
prevalence of diabetes. Although the age group was same, however exposed and unexposed
groups were not similar based on education and energy intake components. This might influence
the outcome of the study.

4EVIDENCE BASED NURSING RESEARCH
The lost-to-follow up in the study was balanced by ensuring that each participants
contributed person-time from January 1998 till date of type 2 diabetes diagnosis. Hence, these
groups were similar to the rest of the cohort.
Measurement error:
To evaluate link between exposure and outcome, it is necessary to blind assessor to
prevent measurement errors. However, no such consideration was taken in research contributing
to measurement errors.
The study aimed to establish links between exposure and outcomes and not evaluation
effect of intervention on a population group. Hence, no systematic differences in care were
analyzed.
Error in research can be systematic or differential (random). Systematic error might be
present in the study because exposure to contaminants for all types of fish was not assessed. The
categorization was done on the basis of serving and not the degree of contamination in each type
of fish. The fish consumption was categorized into different quinitiles and servings per month.
However, fish consumption based on level of contamination with PCB was not observed.
Confounding
One confounding factor in research is the lack of blinding method to strengthen the
data analysis process. This might have affected the interpretation of data. There is no
confounding in terms of study design as it had the right population group and rigorous criteria for
inclusion of participants in the research study. Many validated assessment tools like food
frequency questionnaire (FFD) were also taken which enhanced credibility of the work. FFQ was

5EVIDENCE BASED NURSING RESEARCH
an evidenced based tool which was tried and tested in many pat research evidence. Hence,
confounding factor was present in the analysis process and not the sample recruitment stage.