• extraversion;
  • living alone;
  • mental health;
  • NEO personality inventory;
  • neuroticism;
  • smoking;
  • university students


  1. Top of page
  2. Abstract

Aims:  Frequent onset of several mental disorders starts around undergraduate age for university students. Mental health check-ups of the new students might help provide them with useful supports for improving their mental health. However, few studies have examined the validity of the check-up methods.

Methods:  Whether the scores of a five-factor personality inventory (NEO-FFI) at matriculation predict the needs of mental care and treatment during the first year of the undergraduate course were examined in 8287 new students of a university in Tokyo.

Results:  Logistic regression showed that high neuroticism, low extraversion and high openness of NEO-FFI, majoring in literature/philosophy/ psychology and living out of home were associated with need for mental care/treatment, in addition to the previous use of mental care services.

Conclusions:  Personality inventory such as five-factor ones may be a useful supplemental tool for mental health check-up at matriculation to predict future needs of mental support in undergraduate university students. Students who smoke, live alone out of home and major in subjects such as philosophy might need to be more carefully supported than other students.

NEW UNDERGRADUATE STUDENTS are at the right age for frequent onset of mental disorders, including mood disorder, anxiety disorder, schizophrenia and others. For example, mean age of onset of schizophrenia is around the low 20s in men and the late 20s in women.1 In major depressive disorder frequent onset occurs around the middle 20s.1 A peak of onset of major depression is observed at adolescence, while the mean age of total onset is around age 40.1 The majority of the students leave the protective environment of home and start life by themselves when they enter university. The start of new life may be stressful for the students and could be a trigger for onset or exacerbation of mental health problems.2 Examination of mental health at matriculation may therefore be of great use, which might enable new students to receive appropriate and early mental health care and support.

For this purpose, most of the universities in Japan give mental health check-ups to their new undergraduate students at matriculation. The check-up methods are, however, varied among the universities. In addition, few attempts have been made to evaluate the effects of the check-ups on prediction or prevention of onset and exacerbation of mental disorders. At University of Tokyo a short psychiatric interview is given to all new undergraduate students as a screening for the needs of mental support and care. At the interview, NEO-FFI,3 which is the short version (with 60 items) of the five-factor model personality inventory, NEO-PI-R,3 was used as a supplemental tool for five years. The five factors of the NEO personality inventory are Neuroticism, Extraversion, Openness, Agreeableness and Conscientiousness.

The purpose of the present study was to improve the check-up and to make more effective preventive use of the check-up data for earlier support of students with mental health problems. Especially, we focused on evaluation of usefulness of the NEO-FFI and other questions in the mental health check-up for detection and prediction of mental disorders in new students. Whether the NEO-FFI scores at matriculation were associated with diagnosis with mental disorders during the first year of the undergraduate course was studied. Thus far, a number of studies have investigated the validity of the five-factor model of personality in mental disorders.4–6 These studies found significant associations between NEO-FFI scores and mental disorders, including depressive disorders,7,8 anxiety disorders,4 obsessive-compulsive disorder9 and schizophrenia.7 Variables such as living conditions and the course (or subjects) that the students majored in, and others were also examined.


  1. Top of page
  2. Abstract

During the 3-year period (2002–2004) a total of 9777 students entered the university and all students received the health check-ups. The check-ups were conducted under the order of the president of the university, according to the School Education Law of Japan. For these years, 100% of the new students received the check-up, including the mental health check-up. Foreign students were not asked to fill in the NEO-FFI due to language problems (n = 139). A small number of Japanese students (n < 10) did not fill in the NEO-FFI because of either a lack of time or refusal. The present analysis was conducted in the 8287 students (6616 males and 1671 females; 84.8% of the total 9777; age, 18.5 ± 1.2 years (mean ± S.D.)). Subjects with missing data were excluded. The missing data was on NEO-FFI (n = 290), on living condition (n = 722), on drinking habits (n = 566), on smoking habits (n = 329) and others.

For matriculation into the undergraduate course of the University of Tokyo, students choose one of three courses of humanities or three courses of sciences at the entrance examination. The humanities courses consist of those that approximately correspond to law (the H1 course), economy (H2) and literature–history/psychology/philosophy (H3). The science courses consist of those that approximately correspond to biological sciences (the S2 course), medicine (S3) and other natural sciences/technology (S1). During the years 2002–2004, 4183 new undergraduate students (or 42.8%) entered the humanities courses and the rest (57.2%) entered the science courses in the University of Tokyo.

The health check-up at matriculation consisted of several physical check-ups and a mental health check-up. The mental health check-up was conducted to give preventive mental support or advice to the new students who need the support. A short interview (approx. 3 min on average) was given by psychiatrists (or psychologists) using questionnaires including questions on sociodemographic variables and an 18-item questionnaire on present and past mental status. The Japanese version of the NEO-FFI for university students10 was used as a supplemental tool, in addition to the questionnaire. The sociodemographic variables in the questionnaire included age, gender and number of siblings. Whether the student is going to live at home with their family or not was also asked. The 18-item questionnaire concisely asks about experiences of several mental symptoms, including depressive feelings, idea of suicide, obsessive conduct, panic attacks, hallucinations and others, based on DSM-IV criteria and previous use of counseling or psychiatric clinics. Smoking and drinking habits are also asked about in the questionnaire. Answering the questionnaire takes 5–10 min for average students. The NEO-FFI consists of 60 self-report items on a 5-point scale (from ‘strongly disagree’ to ‘strongly agree’). The NEO-FFI has been used widely in psychobiological studies.11,12 The Japanese version has also demonstrated good internal consistency, test–retest reliability, and validity.13 The NEO-FFI takes 10–15 min to complete for average students. The new students are given the option of refusal; it is written on the questionnaire sheet that the check-up is used for the support and care of their mental health but the answers may be statistically analyzed for the preventive studies. A few students a year refused to answer the questions.

Statistical analysis

Whether the NEO-FFI scores predict diagnosis of mental disorders during the first year of the undergraduate courses was studied. The explanatory variables consisted of the NEO-FFI scores, sociodemographic variables including age, sex, living with family or not after matriculation, undergraduate course during the first and second year (H1–3, S1–3), drinking habit, smoking habit, and the past use of psychological counseling or psychiatric clinics.

The dependent variable was whether the student was diagnosed with any DSM-IV mental disorder or not, during the first year of the undergraduate course. When a student visited the Health Service Center of the university because of problems that started before or after matriculation, and was diagnosed with a mental disorder by a psychiatrist during the first year of the undergraduate course, the dependent variable was determined as 1. The 1 was assigned regardless of whether or not the student had been recommended to receive the more intensive check-up at the matriculation check-up. Otherwise, 0 was assigned as the dependent variable. Students with a score of 0 included those who visited the Health Service Center, with or without the recommendation at the matriculation check-up, but who were not diagnosed with any mental disorder. The Health Service Center of the University of Tokyo provides students with several health-care services such as the health check-ups and medical treatment. Approximately 28 000 students, including 13 000 undergraduate students, are studying in the University of Tokyo, and 700–800 students a year visit the Department of Psychiatry of the Health Service Center for their mental problems. For the statistical analysis, stepwise logistic regression (with step-up procedure and stepwise method) was conducted using SPSS version 12 (SPSS, Chicago, IL, US).


  1. Top of page
  2. Abstract

The average t-scores of six scales of NEO-FFI were 42.7 (SD = 9.9) for neuroticism, 49.4 (SD = 9.2) for extraversion, 50.4 (SD = 9.4) for Openness, 50.7 (SD = 9.8) for Agreeableness and 54.7 (SD = 10.1) for Conscientiousness in the total students at the check-up. Approximately one-half (50.5%) of the total new students were going to live at home (mostly with the parents) and the rest out of home.

During the first year of the undergraduate course,8287 undergraduates, 271 (or 3.3%) out of the 8287 total new students, visited the Department of Psychiatry of the Health Service Center for their subjective mental problems, with or without the recommendation at the matriculation check-up and received the examination by psychiatrists. Out of the 271 students, 203 (74.9% or 2.4% of the total new undergraduates) were diagnosed with a mental disorder. The rest were diagnosed without any mental disorders. The diagnoses in the 203 students were as follows: psychotic disorders including schizophrenia, n = 13; bipolar disorders, n = 9; other mood disorders such as major depression and dysthymic disorder, n = 46; anxiety disorders, n = 30; adjustment disorder, n = 24; one of the three sleep disorders, n = 13; personality disorders, n = 17; and others including tentative diagnoses of Axis I or II disorders, n = 43. Sixty-eight were not diagnosed with any mental disorder. These included 59 students who were invited to visit the Health Service Center at the check-up because of their worry about new life, difficulty in waking up in the morning or other issues, but no symptoms or problems in their adjustment were observed at the clinical interview.

Results of the logistic regression analysis are shown in Table 1. Previous use of mental health services was the most significant (95% confidence interval: 4.82–10.90), which was followed by smoking habits (1.13–5.50), higher neuroticism (1.87–2.59), lower extraversion (0.54–0.76), selection of the H3 course (1.06–2.12), living out of home (1.09–1.99) and higher openness (1.17–1.61). The odds ratios of the NEO-FFI factors are for 10-point increase of the scores. Smoking habits was present in 1.6% of the new students (n = 152). Regarding the course, we had a strong clinical impression that the students who were majoring in or interested in the subjects such as philosophy, literature psychology or related subjects, which correspond to the H3 course, might have more anxiety or emotional instability. In the statistical analysis, the six courses (H1–3, S1–3) were first tested as one categorical variable in a logistic regression model, which was not significant in the analysis. We then replaced the variable with an alternate one ‘majoring in philosophy, literature, psychology or related subjects (H3 course) or not’, according to our strong clinical impression. Those six variables explained 20% of the variance for being diagnosed with a mental disorder during the first year (Nagelkerke R2 = 0.20) in the new undergraduate students (Table 2).

Table 1.  Subject data for 8287 new undergraduate students
  1. NEO-FFI, NEO Five-Factor Inventory.

Age (years) (mean ± SD)18.5 ± 1.2
 n (%)
Male6616 (79.8)
Female1671 (10.2)
Living out of family home4186 (50.5)
Drinking habits (+)105 (1.3)
Smoking habits (+)138 (1.7)
Previous use of mental health services206 (2,5)
Diagnosed with a mental disorder during the first year in the health service center203 (2.4)
T-scores of NEO-FFI (±SD)Mean (S.D.)
Neuroticism42.7 (9.9)
Extraversion49.4 (9.2)
Openness50.4 (9.4)
Agreeableness50.7 (9.8)
Conscientiousness54.7 (10.1)
Table 2.  Variables significantly associated with mental disorder diagnosis during first year university
VariablesβWald χ2OR95%CI
  • CI, confidence interval; OR, odds ratio;

  • Nagelkerke R2 = 0.20; selection criteria of stepwise = 0.05.

  • Odds ratios for 10-point increase of the T scores.

Previous use of mental health services1.9890.977.254.82–10.90
Living out of family home0.396.401.471.09–1.99
Majoring in philosophy, literature, psychology or related subjects0.415.271.501.06–2.12
Smoking habits0.915.152.501.13–5.50


  1. Top of page
  2. Abstract

The present study showed that neuroticism, extraversion and openness scores at matriculation may be associated with being diagnosed with a mental disorder during the first year of undergraduate study. Among the three factors of personality, neuroticism was the most significantly associated and low extraversion was the next. When the T score of neuroticism was increased by 10 points, the risk of being diagnosed was more than twice increased. With the decrease of the extraversion score by 10 points, the risk was increased approximately 1.5-fold. These suggest that the use of the five-factor inventories such as NEO-FFI at matriculation may help predict the need for mental health support in new university students. The personality inventory might be a useful tool at the matriculation mental health check-up. Previous studies observed the association of high neuroticism with depressive disorders.4,8,9,14 Low extraversion was also observed to associate with depressive disorder,5 social anxiety and agoraphobia.4 Low neuroticism and high extraversion was observed to augment resilience against stressor.15 Also, a study reported an association between high openness and obsessive-compulsive disorder.9 The present result is consistent with these previous observations.

Among the explanatory variables, previous use of mental health services was the most significantly associated with being diagnosed with a mental disorder, which was as expected. Living out of home was also significantly associated with this. This is consistent with a previous observation that leaving home, without family or friends, might be associated with depressive disorders in students.2 A marked change in living environment and a decrease of social support might cause this effect.

Smoking habit was also significantly associated with diagnosis of mental disorder. A previous study observed that smokers are more likely to develop mood disorders than non-smokers.16 Other studies observed that smokers had high neuroticism17 or higher openness.18 Symptoms of panic disorder may be more improved when the patients quit smoking.19 These may be related to the present observation. Another significant variable was the selection of the H3 (literature–history/philosophy/psychology) course. Students who major in those subjects could be more vulnerable to mental disorders than students who are majoring in sciences or more practical subjects.

A major limitation of the present study was the possibility that not all the students visited the university health service center when they needed care or support for their mental health. A portion of the students could have used other psychiatric clinics or hospitals. However, the number of such students might not be substantial compared with those who visited the Health Service Center. During 2002–2004, 3.3% of the new students visited the Department of Psychiatry of the Health Service Center and 2.4% were diagnosed with a mental disorder. In contrast, a community study observed that the rate of use of psychiatric services was 1.2% in Japanese adolescents and adults.20 For this reason we may assume that the presence of the students who used other clinics or hospitals did not significanty affect the results of the logistic regression in the present model of analysis, because their number must be very small compared with students who did not have any need for psychiatric sevices.

Another limitation is that the University of Tokyo is one of the few top universities in Japan. Therefore, it is not clear whether the present result can be generalized to new students of other universities. For example, some of the NEO-FFI scores were different for the new students at the University of Tokyo compared to those of average university students in Japan. At the present check-up the university student version of the NEO-FFI was used and the mean T scores of neuroticism and conscientiousness were 42.7 and 54.7, respectively. This means that neuroticism was lower and conscientiousness was higher in the new students at the University of Tokyo, compared to average university students in Japan. Lower anxiety or higher confidence and continuous hard work may be required to pass the entrance examination for the top university. The present results should be interpreted with caution regarding this aspect of the present subjects.

It should also be noted that the six variables explained only 20% of being diagnosed with a mental disorder. The method of the check-up should therefore be improved to make it more sensitive and useful.


  1. Top of page
  2. Abstract
  • 1
    American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, 4th edn, Text Revision; DSM-IV-TR. American Psychiatric Association, Washington, DC, 2000.
  • 2
    Stroebe M, Van Vliet T, Hewstone M, Willis H. Homesickness among students in two cultures: Antecedents and consequences. Br. J. Psychol. 2002; 93: 147168.
  • 3
    Costa PT Jr, McCrae RR. NEO PI-R Professional Manual. Psychological Assessment Resources, Odessa, FL, 1992.
  • 4
    Bienvenu OJ, Brown C, Samuels JF et al. Normal personality traits and comorbidity among phobic, panic and major depressive disorders. Psychiatry Res. 2001; 102: 7385.
  • 5
    Bagby RM, Bindseil KD, Schuller DR et al. Relationship between the five-factor model of personality and unipolar, bipolar and schizophrenic patients. Psychiatry Res. 1997; 70: 8394.
  • 6
    Bagby RM, Joffe RT, Parker JDA, Levitt AJ, Kalemba V, Harkness KL. Major depression and the five-factor model of personality. J. Pers. Disord. 1995; 9: 224234.
  • 7
    Petersen T, Bottonari K, Alpert JE, Fava M, Nierenberg AA. Use of the five-factory inventory in characterizing patients with major depressive disorder. Compr. Psychiatry 2001; 42: 488493.
  • 8
    Duggan C, Sham P, Lee A, Minne C, Murray R. Neuroticism: A vulnerability marker for depression evidence from a family study. J. Affect. Disord. 1995; 35: 139143.
  • 9
    Bienvenu OJ, Samuels JF, Costa PT, Reti IM, Eaton WW, Nestadt G. Anxiety and depressive disorders and the five-factor model of personality: A higher- and lower-order personality trait investigation in a community sample. Depress. Anxiety 2004; 20: 9297.
  • 10
    Shimonaka Y, Nakazato K, Gondo Y, Takayama M. NEO-PI-R, NEO-FFI Manual for the Japanese Version. Tokyoshinri, Tokyo, 1999 (in Japanese).
  • 11
    Lang UE, Hellweg R, Kalus P et al. Association of a functional BDNF polymorphism and anxiety-related personality traits. Psychopharmacology 2005; 180: 9599.
  • 12
    Sen S, Nesse RM, Stoltenberg SF. A BDNF coding variant is associated with the NEO personality inventory domain neuroticism, a risk factor for depression. Neuropsychopharmacology 2003; 28: 397401.
  • 13
    Shimooka J, Nakazato K, Gondo Y. NEO-PI-R and NEO-FFI in Japanese Version. In : TsujiH (ed.). Theory and Practice for the Five-Factor Personality Inventory. Kitaojishobou, Kyoto, 1998; 4759.
  • 14
    Chopra KK, Bagby RM, Dickens S, Kennedy SH, Ravindran A, Levitan RD. A dimensional approach to personality in atypical depression. Psychiatry Res. 2005; 134: 161167.
  • 15
    Campbell-Sills L, Cohan SL, Stein MB. Relationship of resilience to personality, coping, and psychiatric symptoms in young adults. Behav. Res. Ther. 2006; 44: 585599.
  • 16
    Breslau N, Novak SP, Kessler RC. Psychiatric disorders and stages of smoking. Biol. Psychiatry 2004; 55: 6976.
  • 17
    Terracciano A, Costa PT. Smoking and the Five-Factor Model of personality. Addiction 2004; 99: 472481.
  • 18
    Yoshimura K. The psychological characteristics of tobacco dependence in a rural area of Japan. J. Epidemiol. 2000; 10: 271279.
  • 19
    Breslau N, Novak SP, Kessler RC. Daily smoking and the subsequent onset of psychiatric disorders. Psychol. Med. 2004; 34: 323333.
  • 20
    Kikkawa T. Kokorono Kenkouni Kansuru Ekigaku Cyousano Jisshihouhouni Kansuru Kenkyu. Research Report of grants from the Health, Labour and Welfare Ministry of Japan. Ministry of Health, Labour and Welfare, Toyko, 2004 (in Japanese).