N508 Module 3 assignment: research design and epidemiology paper, a full sample

From Counting Cases to Testing Causes: An Emerging Infectious Diseases Study, Descriptive and Analytic Epidemiology, and the Protection of Participants

Student Name

Master of Science in Nursing Program, Aspen University

N508: Theory and Research

Instructor Name

Month Day, Year

From Counting Cases to Testing Causes: An Emerging Infectious Diseases Study, Descriptive and Analytic Epidemiology, and the Protection of Participants

Epidemiology is the study of how disease is distributed in populations and what determines that distribution. Its designs range from descriptive studies that count and characterize cases to analytic studies that test hypotheses about causes. Nurses use both kinds of evidence, often without naming them, when they read an outbreak report or a study of risk factors. This paper analyzes a descriptive study published in the Centers for Disease Control and Prevention's journal Emerging Infectious Diseases, contrasts it with an analytic case-control study from the same journal, and reviews how research participants are protected in epidemiologic work, where consent is often impractical.

The Descriptive Study: Candida auris Hospitalizations

Candida auris is a multidrug-resistant yeast that spreads in health care facilities and can cause invasive infections with high mortality. Benedict et al. (2023) used a large U.S. hospital discharge database to describe hospitalizations associated with C. auris from 2017 through 2022. They identified 192 such hospitalizations, 38 of them (20 percent) involving bloodstream infections. Patients had extensive concurrent medical conditions and heavy prior health care use, and the estimated crude mortality rate was 34 percent. The authors concluded that the findings support continued surveillance and containment efforts.

The study is descriptive: it characterizes who was affected, where, and with what outcomes, without testing a hypothesis about causes. Its strengths are its national scope and its use of a large database that captures clinical detail, and it answers practical questions for infection preventionists and nurses, such as what kinds of patients are affected and how often infection invades the bloodstream. Its limitations are those of descriptive designs. With no comparison group, it cannot show which characteristics increase the risk of C. auris; a database of participating hospitals may not represent all U.S. hospitals; and crude mortality reflects patients' severe underlying illness as well as the infection itself, so it cannot be read as the death rate caused by C. auris. A descriptive study tells nurses what the problem looks like; it cannot tell them what causes it.

The Analytic Study: Risk Factors for Hospitalization With Pandemic Influenza

Analytic epidemiology tests hypotheses by comparing groups. Ward et al. (2011) conducted a case-control study in Sydney, Australia, during the 2009 influenza pandemic to identify risk factors for hospitalization with pandemic H1N1 infection among people older than 16 years. They compared 302 case-patients who were hospitalized with 603 controls from the community and used logistic regression, adjusting for age and sex. Pregnancy was the strongest risk factor, with an odds ratio of 22.4, followed by pre-existing lung disease, immune suppression, asthma requiring regular preventive medication, diabetes, heart disease, and current or former smoking. Obesity was not independently associated with hospitalization but was associated with requiring mechanical ventilation.

The case-control design is efficient for identifying risk factors, especially when an outcome is uncommon, because investigators start with people who have the outcome and look back at their exposures. Its strengths here include a clear case definition, a control group twice the size of the case group, and adjustment for confounding by age and sex. Its limitations include possible recall bias, since exposures were reported after the fact, and the difficulty of choosing controls who represent the population that produced the cases. Odds ratios estimate association, not proof of cause, although a very large odds ratio such as the one for pregnancy, consistent with other studies, strongly suggests a real effect.

The Measures Each Study Uses

The two studies also rely on different measures, and reading them correctly is part of appraisal. Descriptive studies report frequencies and proportions. The C. auris study reports a count of hospitalizations, the proportion with bloodstream infection, and a crude mortality rate, the share of patients who died without adjustment for their other conditions. Such measures describe burden but carry no comparison, so they cannot say whether C. auris patients die more often than similar patients without the infection.

Analytic studies report measures of association. A case-control study reports odds ratios, which compare the odds of an exposure among cases with the odds among controls. An odds ratio of 1 means no association; values above 1 suggest the exposure is more common among cases. The 95 percent confidence interval shows the precision of the estimate: the interval for pregnancy, 9.2 to 54.5, is wide because relatively few participants were pregnant, but it lies far above 1, so the association is unlikely to be due to chance. Cohort studies, by contrast, follow exposed and unexposed groups forward and report relative risks, which are easier to interpret but require much larger or longer studies for uncommon outcomes.

Adjustment is the final concept. Because older people are more likely both to have chronic diseases and to be hospitalized, an unadjusted association between heart disease and hospitalization could simply reflect age. Ward et al. (2011) adjusted for age and sex to reduce that confounding, although factors they did not measure could still play a role.

Comparing the Two Designs

The two studies answer different questions and belong at different points in the investigation of a disease. Descriptive studies come first. They define the problem, generate hypotheses, and guide resource allocation: the C. auris study tells hospitals that affected patients are severely ill and heavily exposed to health care, which suggests where to focus screening. Analytic studies follow, testing whether suspected factors actually increase risk: the influenza study turned the hypothesis that pregnancy and chronic disease increase risk into estimates that guided vaccination priorities. In evidence hierarchies, descriptive studies rank lower than analytic studies because they cannot test causation, but that ranking understates their value; without careful counting, analytic studies would have nothing to test.

For nursing practice, the distinction guides how evidence is used. A nurse reading the C. auris report should use it to understand the burden and to advocate for surveillance and infection control, not to conclude that any particular patient characteristic causes infection. A nurse reading the influenza case-control study can use it to prioritize patient education and vaccination for pregnant patients and those with chronic lung disease, while recognizing that it shows association within one population.

Protecting Research Participants

Epidemiologic research often uses data about people who never agreed to be studied: hospital records, surveillance reports, and databases. The ethical principles that govern research with human participants still apply. The Belmont Report established three principles, respect for persons, beneficence, and justice, expressed through informed consent, assessment of risks and benefits, and fair selection of participants (National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research, 1979). Federal regulations based on those principles require institutional review board oversight of most research with human subjects and allow a board to waive individual consent when three conditions hold: the research poses no more than minimal risk, it would be impracticable to do it any other way, and waiving consent will not harm participants' rights or welfare.

The two studies illustrate different protections. Research using a large, de-identified hospital database, like the C. auris study, poses minimal risk and relies on removal of identifiers and data use agreements rather than individual consent. The influenza case-control study involved interviewing case-patients and controls, so participants were asked for consent and told how their information would be used. Public health surveillance and outbreak investigations conducted by health authorities under their legal mandate are generally treated differently from research, but when their data are later used for research, the same principles of privacy, minimal risk, and oversight apply. For nurses who conduct or assist with epidemiologic research, the key recommendations are to consult the institutional review board early, collect only the data needed, protect identifiers, and respect participants' right to decline when consent is sought.

Justice deserves particular attention in infectious disease research. Outbreaks often fall hardest on people in nursing homes, prisons, crowded housing, or low-wage jobs, who may be studied intensively during a crisis and then see little benefit afterward. Fair research in these settings means involving affected communities in how findings are shared, avoiding language that stigmatizes groups linked to an outbreak, and ensuring that the knowledge gained, such as better screening or vaccination priorities, reaches the populations that made the research possible.

Conclusion

The Candida auris study and the pandemic influenza study show epidemiology's two main modes. The first describes a new threat, revealing a severely ill, heavily treated population with high crude mortality, and the second tests which factors raise risk, identifying pregnancy and chronic disease as strong predictors of hospitalization. Each has strengths and limits that a nurse must understand to apply its findings correctly. Both depend on the trust of the people whose data make them possible, which is why the principles of respect, beneficence, and justice, and the oversight that puts them into practice, remain essential even when individual consent is not.

References

Benedict, K., Forsberg, K., Gold, J. A. W., Baggs, J., & Lyman, M. (2023). Candida auris-associated hospitalizations, United States, 2017-2022. Emerging Infectious Diseases, 29(7), 1485-1487. https://doi.org/10.3201/eid2907.230540

National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research. (1979). The Belmont report: Ethical principles and guidelines for the protection of human subjects of research. U.S. Department of Health, Education, and Welfare.

Ward, K. A., Spokes, P. J., & McAnulty, J. M. (2011). Case-control study of risk factors for hospitalization caused by pandemic (H1N1) 2009. Emerging Infectious Diseases, 17(8), 1409-1416. https://doi.org/10.3201/eid1708.100842