HLTH 6740 Week 5 Disaster Study Evidence Critique Example

Does the Rise Survive the Design? An Evidence Critique of the Katrina Panel Study on Worsening Mental Illness

Student Name

Walden University

HLTH 6740: Disaster, Crisis, and Trauma

Instructor Name

Month Day, Year

Does the Rise Survive the Design? An Evidence Critique of the Katrina Panel Study on Worsening Mental Illness

The Study and Its Claim

Kessler et al. (2008) reported that, among a representative panel of 815 pre-hurricane residents of areas affected by Hurricane Katrina, the estimated prevalence of PTSD rose from 14.9% at the first interview, five to eight months after the storm, to 20.9% at the second, about a year later. Estimates of serious mental illness and of suicidal thoughts and plans rose over the same interval, with suicidal ideation roughly doubling. The authors noted that this pattern ran contrary to most disaster studies, in which distress declines over time, and found that unresolved hurricane-related stresses accounted for large proportions of the increases. This critique asks whether the reported rise survives the design that produced it, without either accepting it uncritically or dismissing it. It works from the published summary of the design and results, and it marks the points where the full paper would need to be checked.

What the Design Does Well

The design has real strengths for its question. It is longitudinal: the same people were interviewed twice, so the comparison is within the panel rather than between two different samples, which removes one major source of error in studies that compare separate surveys. The panel was constructed to represent pre-hurricane residents, including people who had been displaced, rather than only those still living in the area, which matters because displaced residents may differ systematically from those who stayed. And the authors measured stressors as well as outcomes, so they could test whether ongoing hurricane-related stress was associated with the change rather than only reporting that change occurred.

Measurement: Screening Scales Standing In for Diagnosis

The prevalence figures come from screening scales used to generate estimates of diagnostic prevalence, not from clinical diagnostic interviews. Screening scales are practical for large surveys but classify some people differently from clinicians. For the rise to be an artifact of measurement, the scale would need to misclassify differently at the two time points, for example if respondents became more willing to report symptoms on a second interview. That is possible but would not obviously produce a rise of six percentage points; a consistent bias would affect both waves similarly and leave the change largely intact. The measurement concern therefore weakens confidence in the exact levels more than in the direction of change.

Attrition and Who Was Reinterviewed

Any panel loses people between waves. If those who dropped out were doing better, perhaps because they had resettled and moved on, the remaining panel would look worse at the second interview even if no one's symptoms changed. The critique cannot settle this from the summary alone; the question to ask of the full paper is how many participants were reinterviewed, how they compared at baseline with those who were not and whether the analysis adjusted for differences. Until that is checked, attrition is the most plausible alternative explanation for at least part of the rise.

Where the Rise Occurred

The study's subgroup results help. The increases in PTSD and serious mental illness were confined to respondents from outside the New Orleans metropolitan area, while increases in suicidal ideation and plans appeared in both subsamples. An artifact of measurement or attrition would be expected to affect all subgroups in similar ways, so a rise concentrated in one area is more consistent with a real difference in what happened to people there, such as slower recovery assistance or greater unresolved losses. This does not prove the explanation, but it makes a purely methodological account less likely.

The Stress Explanation

The authors attributed much of the rise to unresolved hurricane-related stresses, which accounted for large shares of the increases in serious mental illness, PTSD and suicidality. That explanation fits other research: reviews of disaster survivors have found that secondary stressors after the event are among the most consistent predictors of poor outcomes (Norris et al., 2002). But the association between ongoing stress and worsening symptoms runs in both directions: people whose mental health is declining may also find it harder to resolve housing and financial problems. A panel with two waves can show that stress and symptoms changed together; it cannot by itself establish which drove the other.

How the Finding Fits the Wider Evidence

The rise stands against the usual pattern of decline, and the authors said so. Trajectory research shows that most survivors follow resistant, resilient or recovering courses, with a minority on chronic or delayed paths (Norris et al., 2009). A population-level rise is compatible with that picture if a large share of the panel moved onto a delayed or worsening course under ongoing stress. The study therefore does not overturn the general finding that distress usually declines; it identifies conditions, prolonged unresolved stress after a catastrophic event, under which the usual pattern can reverse.

What a Stronger Design Would Add

Two additions would settle much of the uncertainty. A third wave of interviews would show whether the rise continued, leveled off or reversed as housing and finances stabilized, which would help distinguish a delayed trajectory from a temporary spike. And a direct comparison of reinterviewed and lost participants, on baseline symptoms and stressors, would show whether attrition could explain the change. Neither addition would change the study's central contribution, a warning that the usual decline in distress is not guaranteed, but both would let readers attach firmer numbers to it.

Verdict

The reported rise survives the design in direction, though with less certainty about its size. Its strongest support is the longitudinal design and the concentration of the rise in one subgroup; its main vulnerabilities are attrition and the use of screening scales. The stress explanation is plausible and consistent with the wider literature but cannot be established as causal from two waves. A careful reader should take from the study that post-disaster distress can worsen over the first year and a half when stressors remain unresolved, and should treat the specific percentages as estimates with uncertainty attached.

References

Kessler, R. C., Galea, S., Gruber, M. J., Sampson, N. A., Ursano, R. J., & Wessely, S. (2008). Trends in mental illness and suicidality after Hurricane Katrina. Molecular Psychiatry, 13(4), 374-384. https://doi.org/10.1038/sj.mp.4002119

Norris, F. H., Friedman, M. J., Watson, P. J., Byrne, C. M., Diaz, E., & Kaniasty, K. (2002). 60,000 disaster victims speak: Part I. An empirical review of the empirical literature, 1981-2001. Psychiatry, 65(3), 207-239. https://doi.org/10.1521/psyc.65.3.207.20173

Norris, F. H., Tracy, M., & Galea, S. (2009). Looking for resilience: Understanding the longitudinal trajectories of responses to stress. Social Science and Medicine, 68(12), 2190-2198. https://doi.org/10.1016/j.socscimed.2009.03.043