Keywords

1 Introduction

The health of the crews had a great influence on the safety of a long term voyage, so it was necessary to investigate the physiological and psychological change during the whole voyage. Lu et al. [1] adopted a data classification approach to present a computational study, in which they used 41 chemistry indicators to measure seamen’ blood samples collected before and after the sailing, they found that nine of 41 indicators were significantly affected during the sailing.

Some researchers had demonstrated that people show physiological adaptation to the environment. For example, seasickness was a common phenomenon during a long voyage. LI et al. [2] investigated that the habituation of the seasickness occurred in a long voyage, the prevalence and degree of seasickness started to decrease after two weeks. The result inevitably raised the question of whether it was possible that the habituation occurred in some aspects of psychological functioning. Xia et al. [3] examined the psychological and physiological effects on Chinese female seamen before, during, and after long navigation, they found that the scores of SCL-90 psychological evaluation of somatization, anxiety, paranoia were significantly higher following navigation than before the long voyage. This voyage was the first time for female seamen to go to sea that lasted for 4 months.

During the long-term voyage, the sailors underwent enormous mental load because of their special working environment and tense tasks. They were liable to undertake huge psychological stress under such special environmental conditions at sea. Zhang [4] and Zhang et al. [5] reported the general state of Chinese seamen’s mental health was worse than the general population. A self-administered questionnaire was adopted by Kamada et al. [6] seven times to collect the data before and during the voyage, it reported that complaints of neurotic symptoms and fatigue were reported to be much greater during the 40-day voyage than those in a general occupational setting, and “quality of sleep” was significantly correlated with neurotic symptoms for the crews. Jirásek and Hurych [7] proposed a five-factor model to exam spiritual health in a proper framework. These five factors were the authentic mode of existence, relationship with other people, understanding of the world of nature, question of the purpose and meaning of life, transition and transcendence. Yu et al. [8] proposed that long sailing has a bad influence on the mental health of crews, and they found the total score of SCL-90 was mainly related to sleep, fatigue, working pressure, mental quality and ship adaptability based on multivariate linear regression analysis. However, the psychological change of crews didn’t get the attention as it should deserve. This research mainly explored the correlation between the mental health of ocean seamen and the voyage to investigate the effects of a long voyage on their mental health.

Effective measures could be taken to intervene in the crew’s emotional changes. The controlled study of Norris and Weinman [9] has shown statistically significant improvements in self-esteem and coping in the long sail training voyage. In their study, volunteers completed various psychometric questionnaires before and after the transatlantic voyage, the whole voyage takes 3 months and the longest time between ports was 26 days in this case. There was a trend toward reduction in psychological distress after the long sail training voyage.

There were too many disadvantageous environmental factors which would influence mental health, for example, the narrow in closed space, noise, radiation, air pollution, dissimilar cycle of day and night et al. And there are physiological factors mariners need to face, for instance, disturbances in biological rhythm, sleep disorder, social isolation and barrier of personal communications et al. These factors would cause mental stress to mariners during a long voyage, severely influenced their psychological state, increased anxiety and depression, induced cognitive competence and operational performance, increase their error rates of decision-making and operations, and even cause more serious consequences. However, the previous studies rarely concern and control these environmental and social factors. Therefore, it was necessary to explore how did the long voyage affect the mental status in the experimental environment and construct a time prediction model of the mental status during a long voyage.

2 Methods

2.1 Experimental Design

A before-after experimental study and comparisons among homogeneous groups were used to investigate their mental health status before and after a long simulated voyage. 12 people were randomly enrolled to participate in this experiment. Participants filled in the SCL-90, the MMPI-2, and the PANAS in accordance with a procedure that has been approved by the local Ethics Committee [Certificate No: 2019013]. They were equally divided into three groups to live in a simulated ship. This ship provided narrow and closed space, and social isolation to simulate the environment during a long voyage.

2.2 Participants

The study recruited 12 Chinese male seamen that live in a simulated ship with several functional cabins. The 12 male crewmembers were selected in compliance with the standards of the study subjects. Participants were between 24 and 37 years-old and no drug dependence, no infection history, no mental illness, no hereditary diseases, no history of severe allergies, no chronic diseases that require sustained medication, no neurogenic or functional lesions. They also did not have any sleep disturbance nor receive psychotherapy.

2.3 Measures

Such as Symptomchecklist-90 (SCL-90), Minnesota Multiphasic Personality Inventory-2 (MMPI-2), were adopted to evaluate their subjective experience in anxiety and depression, self-rating somatic symptoms and sleep, fatigue and stress status respectively before and after a long voyage. The positive effect and negative affect schedule (PANAS) were used to investigate the change of positive and negative emotions during the whole experiment period. All surveys had no time limit.

SCL-90.

The Symptom Checklist (SCL-90) [10] was a go-item self-report inventory. It was composed of 90 questions, including 10 sub-scales such as somatization, obsessive-compulsive, symptom, depression and paranoia. Because of its convenient checklist format and its multidimensionality, it had been adopted in many scientific studies since it was originated in the U.S. in the early 1970s. Hafkenscheid [10] believed that SCL-90 was a valuable instrument to detect mental states for the general population. An online test was conducted to the volunteers before and after the simulated voyage.

Minnesota Multiphasic Personality Inventory-2 (MMPI-2).

The MMPI-2 [11] is the widely researched instrument as a multidimensional self-report inventory of psychopathology. Williams et al. [11] reported that one of the most valuable effects of the MMPI-2 was to provide information about whether the examinee cooperated with the test administration. The mainstay of the MMPI-2 were the empirically derived Clinical Scales: Scale 1 Hypochondriasis (Hs); Scale 2, Depression (D); Scale 3, Hysteria (Hy); Scale 4, Psychasthenia (Pt); Scale 5, Psychopathic Deviation (Pd); Scale 6, Masculinity/Femininity (Mf); Scale 7, Paranoia (Pa); Scale 8, Schizophrenia (Sc); and Scale 9, Mania (Ma); Scale 10, Social Introversion (Si).

PANAS.

Because of its state-like and trait-like formats, the Positive and Negative Affect Schedule (PANAS) [12] became a widely used adjective-based questionnaire to measure positive affect (PA) and negative affect (NA) in diverse research projects. The PANAS comprises 10 items measuring PA and 10 items measuring NA. The whole simulated voyage was divided into 6 stages according to the watch system. The subjects needed to complete the PANAS several times in every stage. The average scores of all subjects were used to evaluate positive emotion and negative emotion.

2.4 Statistical Analyses and Fitting Analysis

The collected data were processed and analyzed by the software of Statistic Package for Social Science (SPSS) 21.0. SCL-90 Questionnaire test and MMPI-2 Questionnaire test were expressed as mean ± SD. The descriptive statistics, one-way ANOVA and the paired-samples T-tests were adopted for this study in statistical approaches, where P < 0.05 was considered statistically significant. As for PANAS tests, two fitting analysis models were applied to explore the rule of PA and NA change, respectively.

3 Results and Discussion

3.1 Mental Health Assessment Based on SCL-90

Table 1 presents SCL-90 information about the sample as a whole (N = 12). The SCL-90 had 10 dimensions, the total score of the pretest was less than that of the posttest, but there was no significant difference between pretest and posttest (P > 0.05). The average score of the posttest had improved 0.02 comparing to the pretest, and the standard deviation of the posttest decreased, which implied the average level improved. The pretest scores of the dimensions in somatization, depression, terror, paranoia, sleeping and eating were less than those of posttest, the pretest scores of the dimensions in compulsion, interpersonal relationship, anxiety, hostility, psychosis were higher than those of posttest, but there was no statistically significant difference between pretest and posttest (P > 0.05).

Any score greater than 2 was considered abnormal. The detection rate of psycho-logical abnormalities in the dimension of sleeping and eating after the long term simulated voyage was 0.42. The detection rate before the long term simulated voyage was 0.08. It indicated that volunteers had a higher rate in the dimension of sleeping and eating after the long term simulated voyage than before. The detection rates of psychological abnormalities in other dimensions of SCL-90 were less than 0.17. Therefore, sleeping and eating was a sensitive factor reflecting the mental health of the crews on a long voyage.

The pretest scores in all dimensions of SCL-90 were not significantly different from the posttest. The reasons were manifold: Firstly, the duration of the long voyage might be too short, the mental health had no obvious change, or the duration of the long voyage might be too long, the volunteers had adapted the simulated environment, the mental health had already recovered to a normal level after the voyage. Secondly, the simulated environment was a lack of emergency conditions and life-threatening situations. However, it was very difficult to clarify which one was dominant.

Table 1. SCL-90 information about the sample (N = 12).
Full size table

3.2 Mental Health Assessment Based on MMPI-2

Table 2 presented MMPI-2 information about the sample as a whole (N = 12). The MMPI-2 could be divided into two parts: 4 validity scales and 10 clinical scales. Based on items included in the test, scores were calculated for both validity scales and clinical scales. The results revealed significant differences on one validity scale and two clinical scales. The results were also presented below in graphical form (Fig. 1).

The Minnesota Multiphasic Personality Inventory-2 (MMPI-2) has been shown as a useful measure of psychopathology in extensive projects. With reference to the recommended criteria of Graham (2006) [13], MMPI-2 profiles were validated. The posttest score of the dimension in lying was significantly less than that of pretest (P < 0.05). Compared with the pretest, the participants had higher F scores on the validity scales in the posttest, while the participants had lower L scores. F scores indicated that there were more manifestations of psychosis after the long term simulated voyage than before. L scores indicated that the volunteers had more psychologically defensive in the pretest than in the post-test in order to participate in the experiment successfully.

The posttest T-scores of the dimensions in depression and Paranoia were significantly higher than those of pretest (P < 0.05). The other posttest T-scores of the clinical scales were higher than those of pretest, but there was no statistically significant difference between pretest and posttest (P > 0.05). In the clinical scale, the participants agreed to more items in all subscales of MMPI-2, indicating that the participants had more neurosis and psychotic symptoms after the long term simulated voyage than before the voyage.

High scorers of the dimensions in depression were often associated with depressed mood, low self-esteem, lethargy, and feelings of guilt. The posttest T-scores of depression scale were significantly higher than pretest (P = 0.03), which demonstrated that the long term voyage adversely affected the mental health of crew members and increased the risk of depression among crew members. It’s worth noting that D scale was also associated with both psychotic disorders and neurotic disorders. Based on the perspective of clinical practice, depression was a clinical symptom. It could occur not only in severe mood, depression and bipolar depression, but also in schizophrenia, and even obsessive-compulsive disorder and anxiety disorder could be accompanied by depression.

Paranoia Scale elevations were usually associated with being suspicious, aloof, guarded, and overly sensitive. The posttest T-scores of the paranoia scale were significantly higher than pretest (P = 0.04), which indicated that the participants may be more likely to externalize blame and hold grudges after the long voyage.

Table 2. MMPI-2 information about the sample (N = 12).
Full size table
Fig. 1.
figure 1

MMPI-2 clinical scales

Full size image

3.3 Mental Health Assessment Based on PANAS

The whole simulated voyage had 6 stages. The subjects needed to complete the PANAS several times in every stage. The average scores of all subjects in every stage were used to evaluate positive emotion and negative emotion during the long term voyage. The results of positive emotion change and negative emotion change during the long voyage were shown in Fig. 2 and Fig. 3, respectively. According to the recorded data, the minimum value of the positive emotions was in the 5th stage during the whole voyage, the minimum value was 18.25. On the basis of the recorded data, the maximum value of the negative emotions was in the 6th stage during the whole voyage, the minimum value is 12.27.

Based on all recorded experimental data of PANAS, the fitting of a polynomial was used to analyze the change of positive emotion and negative emotion during the long simulated voyage. The red lines in Fig. 2 and Fig. 3 were the trend lines of positive and negative emotions, they were obtained by fitting different models. The trend lines projected forward for one stage were figured to show the recovery of positive and negative emotions.

As for positive emotion, the trend line was calculated by the quadratic polynomial interpolation and minimum mean square error fitting. The time prediction model was described by

$$ y = 0.02019t^{2} - 0.8399t + 28.33 + \varepsilon \;\left( {R^{2} = 0.6745} \right) $$
(1)

Based on this fitting model, the score of positive emotion gradually declined in the first 4 stages, gradually increased in the remaining 2 stages, and the minimum score of positive emotion was 19.60. The model predicted that participants still needed almost 2 stages after the long simulated voyage to adjust their positive emotions to the level of baseline.

Fig. 2.
figure 2

Positive emotion change during the long voyage (The red line is the trend line) (Color figure online)

Full size image

In regard to negative emotion, the trend line was calculated by the cubic polynomial interpolation and minimum mean square error fitting. The time prediction model was described by

$$ y = - 0.0001581t^{3} + 0.006199t^{2} + 0.009890t + 10.44 + \varepsilon \;\left( {R^{2} = 0.9204} \right) $$
(2)

According to this prediction model, the score of negative emotion gradually declined in the whole voyage, and the maximum score of negative emotion was 12.10. The model predicted that participants still also needed 2 stages after the simulated long voyage to adjust their negative emotions to the level of baseline.

Fig. 3.
figure 3

Negative emotion change during the long voyage (The red line is the trend line) (Color figure online)

Full size image

Comparing the changes of positive and negative emotions, this study found that the positive and negative emotions did not start to recover synchronously. It’s worth noting that the time of positive emotions recover to the level of baseline after the voyage was very close to the recovery time of negative emotions after the voyage.

4 Conclusion

Three different measurements, SCL-90, MMPI-2, PANAS, were used simultaneously to explore the effects of long voyages on psychological states. These three measurements reveal the change of mental status from different perspectives before, during, after a long voyage.

The pretest scores in all dimensions of SCL-90 were not significantly different from the posttest, but the detection rate of psychological abnormalities in the dimension of sleeping and eating after the long term simulated voyage was very high. Therefore, the sleeping and eating scale in dimensions of SCL-90 could be regarded as a sensitive factor reflecting the mental health of the crews on a long voyage.

Based on MMPI-2, this research caught sight of that the mental behavioral disorders of mariners were mainly expressed as depression and Paranoia, and that long voyages influence their mental health. Because the participants agreed to more items in all clinical subscales of MMPI-2, it indicated that the participants had more neurosis and psychotic symptoms after the long term simulated voyage than before the voyage.

PANAS could be used as an effective tool to measure the mental status of a long voyage. The change of the positive emotion could be described as a quadratic polynomial model, and the change of the negative emotion could be described as a cubic polynomial model. Though the positive and negative emotions did not start to recover synchronously, the time of positive emotions recovered to the level of baseline after the voyage was very close to the recovery time of negative emotions after the voyage.