Table 3 Prevalence of recurrence

An et al. (2022)

(adolescents) [33]

‘Recurrent dysfunction’ trajectory: 91 participants (37%)

vs. recovery trajectory (n = 107, 43.5%) and delayed dysfunction (n = 48, 19.5%)

37%

-

-

Andersen et al. (2014)

(military) [34]

Relieved-worsening: 2.0% (n = 11)

vs. low-stable 78.1%; low-fluctuating 7.5%; mild distress 4.1%; distressed-improving 2.7%; late onset 5.7%

2%

-

Ansell et al. (2011)

(civilian) [35]

Of the 142 participants with PTSD at baseline, 76.8% (n = 102) remitted by Year 7 and of these 34% had relapsed by Year 7

-

-

34%

Armenta et al. (2019)

(military) [36]

‘Relapse’ group: 24.5%

vs. rapid recovery (31.1%), chronic (26.1%), and gradual recovery (18.3%)

-

24.5%

-

Benítez et al. (2012)

(civilian) [37]

Of the 199 participants with PTSD, 38% (n = 62) had a remission by year 5. Of these 62, 29.5% had a recurrence episode by year 5

-

-

29.5%

Berntsen et al. (2012)

(military) [38]

‘Late-benefit group’ (PTSD pre-deployment, which increased slightly during deployment and on return, before dropping to subthreshold levels 3 months after return and then increasing again 7 months after return): n = 14, 4%

‘Strong-benefit group’ (high level of PTSD pre-deployment which decreased during deployment, dropped to subthreshold levels on return, and then increased at both 3 months and 7 months after return): n = 8, 2%

vs. resilient (low levels at all times) (n = 306, 84%); new-onset (n = 14, 4%); ‘mild-benefit group’ (PTSD pre-deployment which decreased to subthreshold levels on return and then slightly worsened between 3–7 months post-return; n = 24, 7%). The mild-benefit group has not been included as an example of recurrence as while their symptoms worsened again after initial decline, they did not appear to have reached threshold levels)

6%

-

-

Boe et al. (2010)

(civilian) [15]

‘Reactivated PTSD’ was observed in 18.8% (n = 9) survivors, of which 4.2% (n = 2) were possible delayed-onset cases. In 5 of the reactivated cases (10.4%) the PTSD episode fulfilled all DSM-IV diagnostic criteria whilst 4 (8.3%) were sub-syndromal cases

vs. 58.3% (n = 28) resilient, 14.6% (n = 7) remitted, 8.3% (n = 4) chronic

Of the 9 cases of reactivated PTSD, 3 had experienced additional traumatic events (meeting the stressor criterion in DSM-IV). Precipitating events (not necessarily meeting the stressor criterion) included accident (n = 2), loss (n = 3), reminder (i.e. event that resembled or was directly associated with the original trauma, n = 2), physical illness (n = 1), or none identified (n = 1)

All of the reactivated cases had additional lifetime psychopathologies: recurrent depression (n = 5), alcohol abuse or addiction (n = 6), phobia (n = 3), panic disorder (n = 2), single depressive episode (n = 3)

18.8%

-

-

Chopra et al. (2014)

(civilian) [39]

Of the 81 participants with baseline PTSD, 9 reported ‘trauma only’ (no reexperiencing symptoms) at 3 months and 2/9 (22.2%) had PTSD again at 6 months.

Of the 81 with PTSD at baseline, 4 reported ‘no trauma’ at 3 months and 2/4 (50%) met PTSD criteria again at 6 months. (Some confusion arises over why participants who had experienced PTSD in the past would then claim they had experienced ‘no trauma’. The authors attribute this to fluctuations in recall of previously experienced traumatic events.)

Therefore overall, of the 81 participants with baseline PTSD, 4 (4.9%) experienced recurrence.

-

4.9%

-

Davidson et al. (2005)

(military and civilian) [40]

Relapses in the placebo group: 50% (n = 15)

Relapses in the fluoxetine group: 22.2% (n = 6)

Estimated relative risk for relapse was 1.55 (placebo) and 0.44 (fluoxetine). Odds ratio was 3.50 for relapse on placebo relative to fluoxetine

-

-

50% (placebo group)

22.2% (fluoxetine group)

DenVelde et al. (1996)

(military) [41]

13.8% (n = 17) with PTSD onset between 1944–1950 and 35.8% (n = 44) with PTSD onset after 1950 experienced ‘remissions and exacerbations’. That is, 61/123 (49.6%) of those with PTSD experienced recurrences.

Other trajectories included chronic progressive (9.6%, n = 12 onset 1944–1950, 12.2%, n = 15 post-1950 onset) and duration of symptoms less than 5 years (8.1%, n = 10 onset 1944–1950, 20.3%, n = 25 post-1950 onset)

-

49.6%

-

Fan et al. (2015)

(adolescents) [42]

‘Relapsing/remitting’ group (PTSD symptoms fluctuating and showing a cyclical course across the follow-up period): 3.3%

vs. resistance (65.3%), chronic dysfunction (7.2%), recovery (20%), delayed dysfunction (4.2%)

3.3%

-

-

Gonçalves et al. (2011)

(civilian) [43]

PTSD trajectories were examined for 63 participants who completed PTSD measures at all time-points.

Intermittent cases (met PTSD criteria at least one time point but not all) accounted for 57% (n = 36); of these, 4 met criteria at T1 and T3, 2 met criteria at T2 and T4, 3 met criteria at T1, T2 and T4, and 4 met criteria at T1, T3 and T4

vs. stable non-cases (never meeting criteria) (30%, n = 19) and persistent cases (PTSD at all timepoints) (13%, n = 8).

57%

-

-

Gross et al. (2022)

(military) [44]

Recurrence rate not reported

-

-

-

Hansen et al. (2017)

(civilian) [45]

Of the 123 employees present during the bomb explosion, 3.3% (n = 4) experienced recovery and recurrence (i.e. PTSD at T1, no PTSD at T2, PTSD at T3)

vs. 65.9% no PTSD at any time; 0.8% PTSD at T3 only; 1.6% PTSD at T2 only; 1.6% at T2 and T3; 8.9% PTSD at T1 only; 4.9% PTSD at T1 and T2; 13% PTSD at all time-points

Of the 814 employees not present during the explosion, 0.2% (n = 2) experienced recovery and recurrence (i.e. PTSD atT1, no PTSD at T2, PTSD at T3)

vs. 93.7% no PTSD at any time; 0.7% PTSD at T3 only; 1.4% PTSD at T2 only; 0.3% at T2 and T3; 2.3% PTSD at T1 only; 0.7% PTSD at T1 and T2; 0.5% PTSD at all time-points

3.3% (those present during explosion)

0.2% (those not present during explosion)

-

-

Hepp et al. (2008)

(civilian) [46]

Of those with a CAPS score of 30 or more at any time (n = 25, 28%):

‘Delayed increase’ group: n = 7 (28%)

vs. initial increase in symptoms (n = 10), initial decrease in symptoms (n = 8)

-

28%

-

Holliday et al. (2020)

(military) [47]

Recurrence rate not reported

-

-

-

Karstoft et al. (2015)

(military) [48]

‘Relieved-worsening trajectory’ (moderate initial symptom level that decreased somewhat during deployment and then increased drastically after deployment): 2.0%

vs. low-stable (78.1%), low-fluctuating (mild symptoms before and after deployment, 7.5%), mild distress (low symptoms before deployment followed by moderate increase after deployment, 4.1%), late onset (5.7%), distress-improving (moderate symptoms pre-deployment which decreased after deployment, 2.7%)

2%

-

-

Liang et al. (2019) [49]

Liang et al. (2021) [50]

(adolescents)

‘Relapsing’ trajectory (high PTSD score at T1 which decreased to a low level until T3 but then increased again at T4): 17.7%

vs. resilient trajectory (74.9%) and recovery trajectory (7.5%)

17.7%

-

-

Madsen et al. (2014)

(military) [51]

‘Relieved-worsening’ trajectory: n = 9/456 (1.9%)

vs. low-stable (n = 359); late onset (n = 29); low-fluctuating (n = 30); distressed-improving (n = 12); mild distress (n = 17)

1.9%

-

-

Markowitz et al. (2018)

(civilian) [52]

Of the 43 participants who responded to treatment and completed follow-up measures, 23 were remitters and 20 were responders; 3 months later, 27 had achieved remission status, 10 had responder status and 6 (14% of the original 43) had relapsed – including 2 responders to interpersonal psychotherapy, 1 responder to prolonged exposure, 1 remitter to relaxation and 1 responder to relaxation

-

-

14%

Martenyi et al. (2002)

(military and civilian) [53]

Fluoxetine/fluoxetine group: 5.8% (4/69) experienced clinical relapse (57 completed without relapse, 1 discontinued due to adverse event, 3 chose to leave study, 4 were non-compliant)

Fluoxetine/placebo group: 16.1% (10/62) experienced clinical relapse (41 completed without relapse, 3 were lost to follow-up, 2 chose to leave study, 6 were non-compliant)

-

-

5.8% (fluoxetine group)

16.1% (placebo group)

Murphy & Smith (2018)

(military) [54]

Of 960 participants, 1.2% were in the response-remit class

vs. 2.7% displayed low start-high response; 27.5% with a resistant trajectory; 22.9% showed high start-high response; 47.5% showed high start-moderate response

1.2%

-

-

Osenbach et al. (2014)

(civilian) [55]

Relapsing/remitting trajectory: 35%

vs. resilience (28%); chronic (27%); recovery (10%)

35%

-

-

Osofsky et al. (2017)

(civilian) [56]

‘Initial declines followed by increasing symptoms’: n = 108, 32%

vs. steep decreasing symptoms (n = 79, 23%); stable high symptoms (n = 56, 17%); stable low symptoms (n = 97, 29%)

32%

-

-

Perconte et al. (1991)

(military) [57]

Does not report how many participants are in the relapsed group

-

-

-

Sakuma et al. (2020)

(civilian) [58]

‘Fluctuating course’ (cyclical course moving above and below the diagnostic threshold): 3.5% (n = 26)

vs. resistance (never showing more than mild distress; 62.7%); sub-syndromal (elevated symptoms below diagnostic threshold; 24.3%); recovery (initially above threshold symptoms but decreasing to normal level over time; 6.3%); and chronic (pronounced symptoms from the onset; 3.2%)

3.5%

-

-

Solomon & Mikulincer (2006)

(military) [59]

Of the 131 participants with a combat stress reaction, a recurrent trajectory was observed in 32/131 (24.4%) participants: PTSD at T2 and T4 (n = 1); PTSD at T1, T2 and T4 (n = 17); PTSD at T1 and T3 (n = 2); PTSD at T1 and T4 (n = 7); PTSD at T1, T3 and T4 (n = 5)

Of the 83 with no combat stress reaction, a recurrent trajectory was observed in 11/83 (13.2%) participants: PTSD at T2 and T4 (n = 6); PTSD at T1, T2 and T4 (n = 2); PTSD at T1 and T3 (n = 1); PTSD at T1 and T4 (n = 2); PTSD at T1, T3 and T4 (n = 0)

24.4% of those with a combat stress reaction

13.2% of those without a combat stress reaction

-

-

Solomon et al. (1987)

(military) [60]

(Note that this was a study specifically of individuals with reactivated PTSD, so there is no comparison group who did not have a recurrence)

‘Uncomplicated reactivation’ (individuals diagnosed as having a combat stress reaction which they had completely recovered from and were symptom-free and then developed full-blown PTSD after being exposed to a battle situation in a subsequent war): 23% (n = 8)

‘Heightened vulnerability: specific sensitivity’ (individuals who succeeded in professional and social functioning despite persistent minor symptoms, for whom stimuli reminiscent of the original trauma retained the power to reactivate symptoms): 51% (n = 18)

‘Heightened vulnerability: moderate generalised sensitivity’ (individuals for whom symptoms of PTSD remained apparent in civilian settings, but continued to serve in the reserves despite this, who experienced intense anticipatory anxiety when orders for the 1982 war were issued): 9% (n = 3)

‘Heightened vulnerability: severe generalised sensitivity’ (individuals who displayed inability to function in any setting and whose conditions worsened upon call-up to the 1982 war): 17% (n = 6)

-

-

-

Solomon et al. (2018)

(military) [61]

“Recovered followed by delayed-onset PTSD”: 11 (3.5%); “recovered followed by delayed-onset followed by recovered”: 1 (0.3%) (unclear what this means)

vs. resilient (52.1%); chronic (1.9%); delayed-onset at T2 (11.3%); delayed-onset at T3 (8.7%); delayed-onset at T4 (3.2%); recovered at T2 (2.2%); recovered at T3 (0.3%); recovered at T4 (0.6%); delayed-onset followed by recovery (14.9%)

(It should be noted that a three-class solution represented the best fit for the data; chronic vs. resilient vs. delayed-onset)

3.8%

-

-

Solomon et al. (2021)

(military) [62]

Between Waves 1–4:

‘Reactivation’ group (participants who initially had PTSD, recovered, then had a reactivation of PTSD at a later measurement): 1.6% (n = 3)

vs. chronic (met PTSD criteria at all waves, 5.4%); delayed (did not endorse PTSD criteria in first wave but did in later wave/s, 35.1%); recovered (endorsed PTSD criteria in the first wave but not in later wave/s, 3.2%); and resilient (never met the criteria for PTSD, 54.6%)

At Wave 5:

16.7% of those who had initially recovered had PTSD again during the COVID-19 pandemic

1.6

-

16.7%

Sørensen et al. (2016)

(military) [63]

Of 384 soldiers with data on trajectories:

Relieved-worsening trajectory: n = 9 (2.1% of the sample)

vs. low stable (77.3%); low fluctuating (7.6%); mild distress (3.9%); distress improving (3.1%); late onset (5.5%)

2.1%

-

-

Sungur & Kaya (2001)

(civilian) [64]

‘Acute recurrent’ group (symptomatic at T1, no symptoms at T2, symptomatic again at T3 and/or T4): 8.9% (n = 7)

vs. no disorder (50.5%); acute PTSD (symptomatic at T1 only; 11.4%); chronic resolved (symptomatic at T1 and T2 only, 1.3%); chronic persistent (symptomatic at all time points, 11.4%); delayed persistent (symptomatic at T2, T3 and T4, 5%); symptoms at 6-months only (1.3%); symptoms at 12 months only (1.3%); symptoms at 18 months only (8.9%)

8.9%

-

-

Zanarini et al. (2011)

(civilian) [65]

Recurrence (defined as meeting the PTSD criteria for one month after a two-year period of remission) occurred in 40% (n = 30) of participants with borderline personality disorder and PTSD at baseline who had experienced a remission (25% by the 4-year follow-up, 36% after both six and eight years, and 40% by the 10-year follow-up). Recurrence also occurred in 6/15 axis II comparison participants who met criteria for PTSD at baseline and had a remission during the follow-up years.

-

-

40%

Zlotnick et al. (1999)

(civilian) [66]

Of 13 participants with lifetime PTSD, 2/13 (15.4%) had a recurrence of PTSD during the follow-up; neither remained in their episode of PTSD during the subsequent follow-up interval

-

15.4%

-