HLTH 8035 Week 4 Study Design Appraisal Example

Design Appraisal: A Study Labeled Longitudinal, a Headline Estimate From One Survey Wave, and What a Cross-Sectional Comparison of Daily Vaping and Chronic Cough Lets Anyone Conclude

Student Name

Walden University

HLTH 8035: Epidemiology: Decoding the Science of Public Health

Instructor Name

Month Day, Year

Design Appraisal: A Study Labeled Longitudinal, a Headline Estimate From One Survey Wave, and What a Cross-Sectional Comparison of Daily Vaping and Chronic Cough Lets Anyone Conclude

Summary of the Study as Presented

The composite study, written in the style of a published report, describes itself as a longitudinal analysis of e-cigarette use and respiratory health in young adults. Its data come from a national-style panel survey that interviews the same respondents every year. Its abstract reports that daily e-cigarette users aged 18 to 25 had higher odds of chronic cough, defined as cough on most days for three months or more in the past year, than respondents who did not vape, with an adjusted odds ratio of 1.9. The discussion concludes that daily vaping is associated with the development of chronic cough. This summary repeats the study's framing without judging it; the sections below judge it.

Design: Read From the Methods, Not the Title

A panel survey can support a longitudinal design, but only if an analysis uses its repeated measures. The methods section shows that this one did not. The headline estimate compares vaping status and chronic cough reported in the same survey wave, the third. Earlier waves were used only to describe the sample's history; no respondent's cough status at an earlier wave enters the model, and no analysis restricts to people free of cough before they began vaping. The analysis is therefore cross-sectional, a single snapshot drawn from inside a panel.

The distinction is not pedantic. Study designs are best classified by how the analysis relates exposure and outcome in time, not by the label of the dataset they come from (Pearce, 2012). A cross-sectional comparison captures exposure and outcome in one snapshot, so it cannot show which came first. Young adults with chronic cough might have taken up vaping, perhaps after quitting cigarettes on a clinician's advice, or might vape more or less because of symptoms. The study's own data could have addressed this, which makes the choice to analyze one wave more consequential.

Population and Comparison

Eligibility covered panel members aged 18 to 25 at wave three who answered both the vaping and the respiratory questions. The report does not say how many eligible members skipped the respiratory section, which matters if people with symptoms were more or less likely to answer it.

The comparison group is the more serious issue. "Non-users" combined never-users of any tobacco product with former cigarette smokers and current occasional smokers. Because cigarette smoking causes chronic cough and is common among people who vape, a comparison group diluted with nonsmokers and mixed with smokers makes the vaping estimate hard to interpret. The adjusted model includes current smoking as a yes-or-no covariate, which cannot capture intensity or years of past smoking. A stronger appraisal target would compare daily vapers who never smoked with never-users of any product.

Measure: What a Prevalence Odds Ratio Describes

The reported measure is a prevalence odds ratio. It describes how much more often vaping and chronic cough occur together in the same people at one time, expressed as odds. It does not describe onset, the rate at which new cases of cough arise among people who vape, which is what "development" in the discussion implies. Chronic cough occurs often enough among young adults that the odds ratio would sit further from 1 than a prevalence ratio would, adding a second reason not to read 1.9 as "nearly twice as likely to develop cough."

Reporting Checked Against STROBE

The STROBE statement lists items that reports of observational studies should include so readers can judge them, such as the setting, eligibility, the timing of measurements, how variables were defined and how missing data were handled (von Elm et al., 2007). Checked against it, the report does well on variable definitions and setting. It falls short on the timing of measurement, since the abstract never says the estimate came from one wave, and on missing data, since nonresponse to the respiratory section is not reported. Complete reporting would have made the design limitation visible sooner. It would not have made the design stronger: STROBE is a checklist for transparency, not a measure of validity.

What a Longitudinal Analysis Would Look Like

For contrast, a prospective study of adolescents and young adults in Southern California followed participants across four survey waves and modeled respiratory symptoms against e-cigarette use over time. It found higher odds of bronchitic symptoms and shortness of breath among recent users, and it showed that the associations weakened after adjustment for cigarette use, cannabis use and secondhand exposure (Tackett et al., 2024). That study used the panel's repeated measures and reported how much of the association depended on co-use, two things the composite study did not do.

What the Evidence Supports

The composite study supports one conclusion: at a single point in time, young adults who vaped daily reported chronic cough more often than a mixed comparison group, after partial adjustment for smoking. It does not support the claim that daily vaping leads to the development of chronic cough, because the analysis cannot place vaping before cough, the comparison group mixes smokers with nonsmokers and the measure describes co-occurrence rather than onset. The finding is a reason to analyze the panel properly, restricted to respondents free of cough at baseline and with never-users as the comparison, not a basis for a causal statement.

References

Pearce, N. (2012). Classification of epidemiological study designs. International Journal of Epidemiology, 41(2), 393-397. https://doi.org/10.1093/ije/dys049

Tackett, A. P., Urman, R., Barrington-Trimis, J., Liu, F., Hong, H., Pentz, M. A., Islam, T. S., Eckel, S. P., Rebuli, M., Leventhal, A., Samet, J. M., Berhane, K., & McConnell, R. (2024). Prospective study of e-cigarette use and respiratory symptoms in adolescents and young adults. Thorax, 79(2), 163-168. https://doi.org/10.1136/thorax-2022-218670

von Elm, E., Altman, D. G., Egger, M., Pocock, S. J., Gøtzsche, P. C., & Vandenbroucke, J. P. (2007). The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: Guidelines for reporting observational studies. PLoS Medicine, 4(10), Article e296. https://doi.org/10.1371/journal.pmed.0040296