Sources of Error in Population-Based Research

Description of the Selected Health Issue (Diabetes) Population (Older Adults) and Practice

Gap (Nutritional Changes)

Diabetes is the seventh leading cause of mortality in the US, and it costs the economy
approximately $327 billion in lost wages and work and medical costs (CDC, 2022). It is
estimated that nearly 1 in 10 people in the US, or 37.3 million Americans have diabetes, and 1 in
5 people do not know they have the disease. Another 96 million or 1 in 3 adults have prediabetes.
Further studies show that nearly 33 percent of adults aged 65 years and over have diabetes. This
population bears the greatest risk of developing heart disease, kidney failure, and hypoglycemia
compared to younger individuals (Endocrine Society, 2022).
Despite older adults bearing the greatest risk of diabetes, new evidence suggests that
making lifestyle changes, especially taking healthy diets (fruits and vegetables), can reduce their
risk of developing the disease or prevent diabetes symptoms from worsening. The problem
(practice gap) is convincing older people to make lifestyle modifications that reflect healthy food
choices and eating. The biggest challenge has been that diabetes is a lifestyle disease caused by
multiple factors, healthy nutrition being just one of them. Some older people might not
exclusively make lifestyle changes because of the uncertainty surrounding the true cause of
diabetes.
How the Treatment of this Population/Issue Could be Affected by having an awareness of

Bias and Confounding in Epidemiologic Literature

Having an awareness of bias (selection and information) and confounding can affect the
treatment of diabetes or diabetic older adults in multiple ways. Most importantly, it will cast
doubt on the proposed treatment modalities’ validity, credibility, authenticity, and reliability

3
(Curley, 2020). For example, awareness of errors made in collecting data used in a study about
nutrition and diabetes in older people (information bias), failure to accurately represent an older
population for whom outcomes of the research were generalized (selection bias), and failure to
consider diabetes confounding factors like smoking and alcohol intake can cast doubt on the
validity and reliability of the proposed recommendations of the study and, ultimately, their
ability to be extrapolated to the larger older adult population.
Two Strategies Researchers Can Use to Minimize Bias and Confounding
There are multiple ways through which a researcher can minimize information bias,
selection bias, and confounding when doing research. For example, a researcher can minimize
selection bias through randomization, in which study participants are assigned by probability or
chance to separate groups (treatment and control). The second way is by ensuring populations,
participants, and subgroups are selected to mirror the characteristics of the larger population
being investigated or surveyed. The second type of bias, Information bias, can be reduced by
implementing standardized protocols/procedures in collecting data across all treatment and
control groups. The second method is by training interviewers or data collectors to use standard
methods when collecting information or data. Finally, researchers can minimize the potential
influence of confounding variables by restricting and matching. Restricting means recruiting
only subjects or participants with similar values of the confounding factors to the treatment
group. Matching means only selecting a comparison (control group) matching characteristics to
the treatment group. Also, during analysis, a researcher can address the confounding effect by
including possible confounding factors as control variables in the regression models (Curley,
2020).
The Effects of These Biases on the Interpretation of Study Results if Not Minimized

4
If not minimized, the two biases (selection and information) and confounding variables
might misrepresent facts and false interpretation of the study outcomes. If generalized, the
information and study results can be misleading and incorrect (Friis & Sellers, 2021).

5

References

CDC. (2022). Diabetes. https://www.cdc.gov/diabetes/library/spotlights/diabetes-facts-stats.html
Curley, A. L. C. (Ed.). (2020). Population-based nursing: Concepts and competencies for
advanced practice (3 rd ed.). Springer.
Endocrine Society. (2022). Diabetes and older adults. https://www.endocrine.org/patient-
engagement/endocrine-library/diabetes-and-older-adults
Friis, R. H., & Sellers, T. A. (2021). Epidemiology for public health practice (6 th ed.). Jones &
Bartlett.