Repeated Assessment of Posttraumatic Stress Disorder Severity and the Risk of Mortality

Strong evidence supports the conclusion that posttraumatic stress disorder (PTSD) is associated with increased risk of numerous adverse health outcomes. As described by Schnurr and Green, poor health and higher mortality rates among those with PTSD may be because of the “downward spiral that follows the wear and tear caused by PTSD.”¹ Existing studies of PTSD and mortality have typically assessed mortality many years after a single baseline PTSD measurment.² This approach does not account for the potential difference in mortality risk among those who have severe, persistent PTSD compared with those who experience remission or those who have a variable pattern of improving and worsening symptoms. Giesinger et al³ performed a large prospective cohort study from the World Trade Center (WTC) Health Registry, using 4 waves of data collection from more than 63 000 civilians and first responders who were exposed to the WTC disaster. Investigators used PTSD Checklist (PCL) scores of 50 or greater to define PTSD and identified mortality through National Death Index data. Among all participants, 2349 deaths occurred during the 13-year follow-up period, including 487 with cardiovascular mortality and 230 with an external cause of death (including suicide and accidental poisoning). The study by Giesinger et al³ was unique in that it compared the magnitude of the association between PTSD and mortality using 2 separate survival modeling approaches. First, PTSD was modeled as present or absent at baseline, and second, PTSD was treated as a time-varying exposure. When PTSD was modeled as present or absent at baseline, the magnitude of the association between PTSD and mortality was smaller and not statistically significant among civilians. When modeled as a time-varying exposure, PTSD was significantly associated with mortality among both civilians and first responders for all-cause, cardiovascular, and external-cause mortality.

The latter modeling approach is a more accurate measure of the association of PTSD with mortality because it allows those who did not have PTSD at baseline to contribute PTSD exposure time if they met the PCL threshold for PTSD at wave 2, 3, or 4 of the WTC Health Registry. In addition, participants could have less severe symptoms at a given wave and would not contribute to the PTSD exposure at 1 or more follow-up waves. This reduced the likelihood that cases were missed or even misclassified during follow-up. Overall, the study demonstrated the total mean association of PTSD with mortality, both in the short-term and in the long-term. However, the treatment of PTSD as a time-varying exposure does not elucidate the association between the course of PTSD and mortality. That is, patients were categorized as having PTSD or not having PTSD at each wave, which does not capture the trajectory of PTSD symptoms over time. Within this sample, it is safe to assume that some participants had persistent severe PTSD (eg, PCL score >70) or persistent moderate PTSD (eg, PCL score 45-55). Others may have had severe PTSD at baseline (eg, PCL score >70) and experienced long-term improvement, while others might have started with a low PCL score that worsened over time. Indeed, separate studies of the WTC Health Registry reported that 36.3% of the sample had nonpersistent PTSD, defined as a PCL score of less than 44 during 1 or more follow-up waves.⁴

More importantly, the study by Giesinger et al³ did not account for the potential benefits associated with meaningful improvement in PTSD. We have reported that a substantial decrease in PTSD symptoms was associated with a lower risk of type 2 diabetes and improvements in several health behaviors.^5-7 In a cohort of veterans treated at the US Department of Veterans Affairs PTSD specialty clinics, we observed that those who experienced a decrease in PCL score of at least 20 points (consistent with a large, clinically meaningful improvement) were 49% less likely to develop type 2 diabetes than those who had a decrease of less than 20 points during a follow-up period of 2 to 6 years.⁵ In separate studies, we observed patients with at least a 20-point PCL score decrease were 37% more likely than those with a smaller decrease to use Veterans Affairs nutrition and weight loss programs,⁶ 78% more likely to be adherent to antidepressant medication,⁷ and 56% more likely to use any substance use disorder treatment (J. Salas, 2019, unpublished data).

Whether clinically meaningful improvement in PTSD reduces the risk of all-cause mortality and cause-specific mortality remains unknown. Teasing out the potential benefit of PTSD improvement on all-cause mortality may be particularly challenging in the WTC Health Registry because first responders and many civilians were exposed to a variety of toxic materials.

The findings of Giesinger et al³ reveal the importance of measuring PTSD symptoms longitudinally. While the study is important and the analyses are robust, future WTC Health Registry research could model the association between PCL improvement and mortality. In addition, a great deal could be learned by determining whether risk of mortality differs among those with persistent, severe PTSD, those with moderately severe PTSD, and those with PTSD remission. Investigators have access to repeated PCL measures over a 13-year period and could compute latent growth trajectories to determine whether differences in the course of PTSD have variable associations with mortality. Although additional research is warranted, the study by Giesinger et al³ is a call to action to identify and treat the consequences of PTSD among individuals exposed to traumatic events.

Published: February 5, 2020. doi:10.1001/jamanetworkopen.2019.20493

Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2020 Scherrer JF et al. JAMA Network Open.

Corresponding Author: Jeffrey F. Scherrer, PhD, Department of Family and Community Medicine, St Louis University School of Medicine, 1402 N Grand Blvd, St Louis, MO 63104 (scherrjf@slu.edu).

Conflict of Interest Disclosures: Dr Scherrer reported receiving compensation for service as editor in chief of Family Practice from Oxford University Press. No other disclosures were reported.

Funding/Support: Drs Sherrer and Schnurr and Ms Salas received support from the National Institutes of Health grant NHLBILR01HL125424.

Role of the Funder/Sponsor: The funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Additional Information: This material is the result of work supported with resources and the use of facilities at the Harry S. Truman Memorial Veterans’ Hospital.

Disclaimer: The views expressed in this report do not necessarily reflect those of the US Department of Veterans Affairs.