As the first group of countries affected by the epidemic, China's psychological dynamic process of the epidemic may occur earlier than that of other countries, which can be used as a reference for other countries. This current study conducted a longitudinal investigation to examine the patterns of PTSD in medical workers in the context of COVID-19. Four trajectories of probable PTSD were formed both in CC group and non-CC group. These results indicate that individual heterogeneity exists in posttraumatic reactions. The results of this study are similar to other research that examine trajectories of PTSD in postpartum women, military members, veterans and earthquake survivors[13, 16, 17], however, uniquely highlight how the symptoms of PTSD move in tandem through time, helps to expand the knowledge about factors involved in the development of COVID-19 related PTSD and distinguish the different PTSD triggers between close contacts and non-close contacts with Coronavirus.
In CC group, a large number of participants (25.28%) demonstrated the resilient trajectory, presenting a stable low level of symptoms, scale scores ranging from 36.35 to 33.96. The recovery trajectory was observed in 36.26% of the present sample; this trajectory is characterized by an initial increase in symptoms followed by a decrease in symptoms to a low level. The two trajectories means that more than half people can finally show very mild or no symptoms of psychological trauma. Studies suggest that there are two main reasons for this finding. First, few deaths and injuries occurred among medical workers. Second, the state and the public provided excellent support and draw great attention to victims of the disease survivors, which might have had positive psychological impacts on these individuals [2]. Our data shows that medical staff who have emergency rescue experience in major public health events are prone to PTSD, but they also recover quickly. Although the variation tendency of PTSD showed by most of the medical staff in CCs is satisfactory, some are not. We have to pay more attention to the chronic PTSD trajectory (16.48%) displaying an stable high level of symptoms and the delayed PTSD trajectory (21.98%) showing delayed symptoms. More than one third of close contacts with COVID-19 have long-term PTSD symptoms. A sustained global epidemic may serving as a continual reminder of the traumatic event, maintaining arousal, and preventing them return to a normal life [9]. The current data demonstrates that the peak scores of the trajectories can reach more than 70 points, which means a severe status. The high rate of chronicity and relatively significant rate highlight the need for continued screening and treatment to promote recovery or prevent the worsening of PTSD symptoms.
Interestingly, Female gender was found to be an independent protective factor for PTSD in CC group, which inconsistent with a previous study conducted from January 29, 2020, to February 3, 2020 [18]. One possible explanation may be the choice of different timing, because the current study was a longitudinal observation involving two time points. As the mental status of medical staff may gradually change over time, women tend to pay more attention to their experiences and feelings and are more willing to express their emotions. This behavior is conducive to the self-regulation of emotions [19]. It carries important implications in the health care system, where the majority of the nursing staff and health care assistants are female. [9] They are better competent at their jobs on the long run. In addition, unmarried individuals shows good psychological adaptability. A recent case control study on healthcare workers facing the COVID-19 pandemic indicates that married individuals get higher scores in vicarious traumatization symptoms compared to unmarried ones [20]. Besides, it is noteworthy that person who worked in high epidemic areas of Covid-19 likely suffer from PTSD (recovery, chronic, delayed PTSD ). A study [21] investigated a sample of 549 medical staff in Beijing (China), finding 2 to 3 times higher PTSD rates among respondents who worked in high-risk locations and perceived high SARS-related risks. Reason for this phenomenon may be related to high level of exposure to atmosphere associated with Covid-19 patients. Close contact and direct exposure to the patients' physical and psychological sufferings have made front-line medical workers prone to suffer from higher risk of traumatization or PTSD [20, 22] .
In non-CC group, it was found the rate of PTSD (34.45%) at 1 month is not as low as we expected and the score (63.63) is very high on the contrary. The incidence and severity were higher than that in previous studies [23, 24].Secondly, non-CCs showed a similar tendency with the CCs that a lot of individuals got chronic or delayed PTSD (22.02%). One-way ANOVA reveals that it was statistically different not only between groups but also changed over time. Thirdly, it found that the participants in major public health emergencies were closely associated with developing recovery PTSD, consistent with CC group. The possible explanation for the initial rising scores of PTSD is that this kind people were more vigilant to the harmfulness of epidemic and more anxious about virus spread losing control than others. Study suggests that perceived threat can be more strongly associated with the PTSD than object exposure[16]. As a result, they could be more prone to be impacted by the negative emotion in the early days. However, being equipped with specialized knowledge, skills and comprehensive handling abilities of emergency enabling them well adaptive to stress, which made them recover soon. Fourthly, the current study determined the senior medical staff was associated with an elevated risk of developing delayed PTSD. As the epidemic continues for more than 1 year, at the stage of regular epidemic prevention and control, hospitals at all grades dedicate to the prevention and control of the epidemic, as well as work resumption and economic development. Compared to the junior level medical staff, the senior level medical staff is under higher pressure on restoring order of society [25]. In light of the above situation, we should highlight the need for continued screening, timely helping and sufficient social supporting [26, 27], aiming to overcome psychological disorder and promote the PTSD recovery.
Comparative analysis conducted between CCs and non-CCs, and there have something in common: ①Both of them have 4 different PTSD symptom trajectories shows extremely similar trend (Fig. 1).②Approximately more than half of the participants(Fig. 3) exposed to the COVID-19 event exhibit acute stress responses immediately and maintain a stable trajectory of euthymia and healthy functioning, which is consistent with the findings of some meta-analyses[28, 29].③Of the current sample, the proportion with chronic and delayed dysfunction over the follow-up period is not low among CCs and non-CCs.
However, CCs was different in severity and influencing factors from non-CCs. Firstly, CCs take higher risk of PTSD. The PTSD incidence in CCs at 1 month (Recovery + Chronic = 52.74% ) was higher than non-CCs(43.45% ), and still remain high level at 12 months (Chronic + Delayed = 38.46% ), which is nearly two times higher than non-CC group (22.02% ) shows in Fig. 3. Thus, CCs are more likely to suffer from acute PTSD, then may develop chronic or delayed PTSD. Besides, 74.73% of CCs are non-resilient and 21.98% of them suffer from delayed PTSD, which is statistically significantly higher than non-CCs. It indicates that close contact with the COVID-19 patients not only increased the possibility of PTSD, but also affected the development of PTSD. Secondly, with the resilient group serving as the reference, bivariate logistic regression shows that the predictive factors have great differences between CCs and non-CCs. Several predictors were found in CC group, while limited found in non-CC group which was not affected by socio-demographic factors. With regard to consistent risk factors, the recovery, chronic and delayed group have been merged to expand the sample size. Then the multivariate logistic regressions find an consistent result with previous analysis of bivariate. The phenomenon of different influencing factors between CCs and non-CCs means that the methods of psychological intervention may need to be differentiated although these two groups share the same symptoms and shows similar trajectory changes. In addition, whether the main predictors of medical staff who non-close contact with COVID-19 patients were other more complex factors remains an important direction for further research.
The study has certain limitations. Firstly, online self-administer and non-rigorous random sampling could have affected the representativeness and reliability of the results. And a certain amount of cases were lost during the second follow-up due to the long observation period. In the current situation, the most important task is to contain the epidemic, and it is very important to reduce people gathering, which cause a major obstacle to this study to carry out large-scale face-to-face investigations. Secondly, more complex factors should be taken into account in the future study, such as personality characteristics or coping skills, etc. The relevant factors which may affect the trajectories of non-CC group from our analysis is limited.