Age at onset of schizophrenia: Gender differences and influence of temporal socioeconomic change

Authors


Correspondence address: SakaeTakahashi Nihon University School of Medicine Department of Neuropsychiatry, 30 Oyaguchi-Kamimachi, Itabashi-ku, Tokyo 173-8610, Japan. Email: sakae@med.nihon-u.ac.jp

Abstract

This study was undertaken to examine whether males develop schizophrenia at a younger age than females, and whether temporal socioeconomic change affects the age at onset of schizophrenia. The subjects were 848 ICD-9 schizophrenics who were admitted to Nihon University Hospital, Tokyo, Japan, during the period of 1955–64 (n = 468 (214 males and 254 females), group A) or during the period of 1982–91 (n = 380 (220 males and 160 females), group B). Schizophrenic males showed an earlier age at onset than schizophrenic females. However, the mean age at onset of schizophrenia did not differ significantly between group A and group B. These results indicate that the gender difference in age at onset of schizophrenia has not been influenced by temporal socioeconomic change.

INTRODUCTION

Since early research on gender differences in schizophrenia by Kraepelin, many studies have been conducted reporting data on this subject. A number of reports have indicated that male schizophrenic patients show an earlier age of onset, a poorer premorbid history, more negative symptoms, a higher relapse rate, a worse outcome, a poorer response to neuroleptic drugs, and a lower family morbidity risk for schizophrenia than female schizophrenic patients. In addition, schizophrenic males have different structural and functional brain abnormalities than schizophrenic females. 1 The ICD-10 also notes that males develop schizophrenia at an earlier age than females. However, no study has been conducted specifically concerning the influence of socioeconomic change over time (temporal socioeconomic change) on age of schizophrenia onset.

We retrospectively investigated the age at onset in two schizophrenic groups consisting of in-patients in the Department of Psychiatry, Nihon University Hospital, Tokyo, Japan. These subjects were selected from two distinct 10-year periods. Primarily, this study was undertaken to examine whether males develop schizophrenia at an earlier age than females, and whether temporal socioeconomic change has affected the age of schizophrenia onset.

MATERIALS AND METHODS

From the records of 2775 psychiatric patients admitted to Nihon University Hospital, which is located in the northern part of Tokyo, 848 patients with schizophrenia in accordance with ICD-9 were selected for retrospective study. Patients were chosen based upon admitting periods during one of two different time periods: between 1955 and 1964 and between 1982 and 1991. All of these subjects were diagnosed with schizophrenia upon admission; however, the original diagnoses were based on a conventional standard. Thus, for the present study, four psychiatrists rediagnosed all ‘schizophrenic’ patients in accordance with the description in the ICD-9 glossary. These particular 848 subjects emerged as meeting the criteria for the present study.

The subjects were divided into two groups. Group A consisted of 468 subjects (214 males and 254 females) who were in-patients in our psychiatric unit during the period between 1955 and 1964, and group B consisted of 380 subjects (220 males and 160 females) who were in-patients during the period between 1982 and 1991. There was no overlap of patients between groups A and B. There was an 18-year time lapse between periods. Over this 18 years, the Japanese population as a whole experienced substantial socioeconomic change. Thus, we consider that temporal socioeconomic change was represented by the two different time periods spanned by the two groups of in-patients. Evaluation of the age at onset was conducted independently by the same four psychiatrists using the hospital charts. The age at onset of illness was defined as the age when initial psychotic symptoms appeared. In the course of evaluating the age at onset of subjects, there were a few discrepancies among four psychiatrists in regard to diagnosis. In these cases, we engaged in serious debate until a consensus was reached.

In the present study, we examined the contribution of temporal socioeconomic change and gender on age at onset of schizophrenia. In order to achieve this, a two-way analysis of variance ( ANOVA) was run for age at onset (dependent variable), and temporal socioeconomic change (the two groups) and gender (as independent variables). Values P < 0.05 were considered significant.

RESULTS

Table 1 shows the average age at onset according to gender and group. A 2 (gender: male and female) × 2 (temporal socioeconomic change: group A (1955–64) and group B (1982–91)) ANOVA revealed only one significant main effect. The effect for gender was significant (F = 3.89, d.f. = (1,844), P value = 0.048): male schizophrenics had an earlier age at onset than female schizophrenics. There was no significant effect for temporal socioeconomic change (F = 0.089, d.f. = (1,844), P = 0.765), and no significant interaction (gender × temporal socioeconomic change, F = 1.02, d.f. = (1,844), P = 0.312).

Table 1.  Mean age at onset for male and female patients in groups A and B Thumbnail image of

Figure 1 shows the frequency of age at onset for male and female patients in groups A and B based upon 10-year increments and for patients 40 years and older. Subjects who developed schizophrenia after the age of 40 were unified into one group because they represented a very small percentage (group A males, 6.5% (14/214); group A females, 7.7% (19/254); group B males, 6.4% (14/220); group B females, 7.9% (20/160)). Onset between the ages of 20 and 29 was more frequent in males than in females. However, onset occurred in females in their 30s and after the age of 40 more frequently than in males of those ages. The trends were constant between groups A and B.

Figure 1.

Frequencies of onset per 10-year age intervals. (–○–), group A male; (–▪–), group A female; (–○–), group B male; (–▪–), group B female.

Gender was demonstrated to influence age at onset of schizophrenia. However, temporal socioeconomic change might not influence age at onset of schizophrenia.

DISCUSSION

A number of studies have observed important gender differences in age at onset of schizophrenia. Angermeyer and Kuhn presented a review of the literature concerning gender differences in age at onset of 36 papers giving data on the onset of schizophrenia in males and females. 2 All but three showed that age at onset of schizophrenia occurred earlier in males than in females. Following Angermeyer’s report, several similar findings were published. 3–7 We also found significant differences between males and females for age at onset of schizophrenia. In addition, we found that the schizophrenic patients who were admitted to Nihon University Hospital, during the periods of 1955–64 and 1982–91 did not show significant differences in mean age at onset. Based on these findings, we hypothesize that schizophrenic men generally do show an earlier age at onset than schizophrenic women, and that gender difference in age at onset is not influenced by temporal socioeconomic change. Lewine et al. indicated that the tendency for men to develop schizophrenia earlier than women has been constant over a considerable span of time, despite changes in sex ratios in the diagnosis of schizophrenia. 8

It is well known that Japan industrialized very rapidly in the postwar era, and that the Japanese economy went through a period of explosive growth beginning in the 1960s. Dramatic change occurred between 1964 and 1983 in industry, society and culture. The standard of living rose greatly, higher education became widely accessible, people married at a later age, and women obtained a higher social status. Moreover, psychiatric practice itself changed during this period with the introduction of treatment medications in the 1950s, increased use of medications during the 1960s, the tendency to treat more patients on an out-patient basis beginning in the 1970s, and the advent of a day care system and rehabilitation for re-entry to society beginning in the 1970s. Therefore, on the basis of our results, we consider that the most important factors regarding gender differences in onset of schizophrenia are biological factors, while temporal socioeconomic change has had little impact on the onset of schizophrenia. In fact, several studies have described biological differences associated with gender in schizophrenia. Some have reported that schizophrenic males have different morphological and physiological brain abnormalities, in addition to a lower family risk, than schizophrenic females. 9–20 Seeman and Lang hypothesized that estrogens account for many of the observed gender differences. 21 They noted that estrogens increase the density of dopamine type 2 receptors as do neuroleptics, and high levels of estrogens may serve a protective function in the onset of schizophrenia. This may in part account for the fact that schizophrenic men show an earlier age at onset than schizophrenic women.

In summary, we conclude that biological factors influence the age at onset of schizophrenia. Furthermore, it may be possible that biological factors influence the age at onset of schizophrenia more potently than social factors. However, we cannot ignore the social factors. In order to verify our results, it is essential to carry out further and more extensive research.

Ancillary