The natural history of psychosomatic symptoms and their association with psychological symptoms: Observations from the Population Study of Women in Gothenburg

Abstract

Objective: To study the prevalence of subjective unspecified psychological symptoms (UPS) in a cohort of middle-aged women and the development of psychosomatic symptoms over 24 years. Methods: In 1968–1969, 1462 randomly selected women, aged 38, 46, 50, 54 and 60, were recruited. A 24-year follow-up was performed in 1992–1993 (n=836). The same structured interview concerning psychological and psychosomatic symptoms was used on both occasions. Results: The prevalence of UPS was 28% in 1968–1969 and 20% in 1992–1993. Women with UPS in 1968–1969 were significantly more likely to have asthma/obstructive symptoms (34% vs 26%), headache (38% vs 22%) and abdominal symptoms (40% vs 21%), but not hypertension/high blood pressure (28% vs 28%), than women who did not report UPS. In 1992–1993, women with UPS in 1968–1969 were significantly still more likely to have asthmatic (25% vs 18%) and abdominal symptoms (44% vs 33%). Sixty-one per cent of women with UPS in 1968–1969 did not report such symptoms in 1992–1993, compared to 86% of women without UPS in 1968–1969 (p<0.001).

Conclusion: Women reporting UPS seemed to have a higher frequency of simultaneous psychosomatic symptoms than women not reporting UPS. However, having UPS was apparently unassociated with the development of psychosomatic symptoms over time. Psychosomatic symptoms in women seem to be self-limiting and decrease with time.

• Longitudinal
• population study
• psychosomatic
• unspecified psychological symptoms
• women

Introduction

Psychosomatics as a concept has been discussed for a long time. Various theories have been put forward over the years to explain psychosomatics, and different hypotheses have been presented to explain the specific area between soma and psyche. So far, no one has been able to provide a complete explanation of how the two are connected [1], [2]. The definition of psychosomatics has also varied over time [3].

Alexander (1950) used the term psychosomatic when pointing out how psychological factors could precipitate or aggravate seven specific diseases, called “the holy seven”, i.e., peptic ulcer, bronchial asthma, rheumatoid arthritis, ulcerative colitis, essential hypertension, neurodermatitis and thyrotoxicosis [4].

Women's health and illness are coming increasingly into focus. Previous studies have shown that Swedish women's self-reported health deteriorated during the 1980s and 1990s, and the prevalence of psychosomatic symptoms such as headache/tiredness and gastrointestinal symptoms has increased, especially in young women [5].

We report from a prospective population study of women, initiated in 1968–1969 in Gothenburg, Sweden, with four additional follow-up examinations in 1974–1975, 1980–1981, 1992–1993 and 2000–2001 [6–9]. Somatic and psychological symptoms and previous and present illness were registered among other variables [6]. In this paper, only data from the 1968–1969 and 1992–1993 examinations are used.

The aim of this paper was to study the prevalence of unspecified psychological symptoms (UPS) and psychosomatic symptoms in an urban female population, and to investigate whether women reporting UPS simultaneously present more psychosomatic symptoms both from a cross-sectional and from a longitudinal perspective. We have chosen to focus on essential hypertension/high blood pressure and bronchial asthma/obstructive symptoms as psychosomatic symptoms from Alexander's original description and have added headache/migraine and abdominal symptoms. We are, however, aware that several other symptoms/illnesses can be an integral part of the psychosomatic symptoms concept.

Participants and methods

Study population

In 1968–1969, 1462 women aged 60, 54, 50, 46 and 38 in Gothenburg, Sweden, participated in the prospective Population Study of Women in Gothenburg. All women examined in 1968–1969 were offered a second examination in 1974–1975. A total of 1302 women attended this examination (89.1% of those originally examined). In 1980–1981, a third examination was conducted on 1154 participants (78.9% of the women originally examined). In 1992–1993, the fourth examination, a 24-year follow-up study took place. The participation rate in 1992–1993 was 57.2% of the original 1462 participants in 1968–1969, 280 (19.2%) of whom had died before the 24-year follow-up study in 1992–1993. This corresponds to a participation rate of 70.2% among those who had participated in 1968–1969 and who were alive in 1992–1993. In Figure 1, the number of women invited to participate and the number who actually participated in the initial and follow-up studies are shown. Details of the sampling procedure and participation rates at all examinations have been presented elsewhere [6–9].

Figure 1.  Flow chart of participants, non-participants and deceased in the Population Study of Women in Gothenburg examinations in 1968–1969 and 1992–1993.

Sampling was based on date of birth, which, together with a high participation rate (90.1%), ensured that participants were a representative cross-section of women from the community in the age groups studied [6].

In this study, women in all age groups (born in 1908, 1914, 1918, 1922 and 1930, respectively) reporting psychosomatic symptoms (hypertension/high blood pressure, asthma/obstructive symptoms, headache, abdominal symptoms) in the 1968–1969 examination were compared cross-sectionally in 1968–1969 and 1992–1993, respectively, and from a 24-year longitudinal perspective (1968–1969 to 1992–1993) with women not reporting such symptoms in 1968–1969. We were able to link cases from the 1968–1969 sample through to the 1992–1993 sample because of the formulation of the questionnaires, which was almost identical throughout the years.

Symptoms and illnesses in the group

The prevalence of UPS was based on a self-administered questionnaire, and UPS were defined as answering yes to the question “Have you had any of the following illnesses/symptoms: psychological/nervous symptoms?” The women who reported having psychological/nervous symptoms were defined as having UPS.

The prevalence of psychosomatic symptoms was based on the same self-administered questionnaire and was defined as answering yes to the question “Have you had any of the following illnesses/symptoms: asthma, headache/migraine or abdominal symptoms (constipation, diarrhoea or other abdominal symptoms)?” Subjects reported whether they had had any of these symptoms. They were interviewed by a physician with the following questions: Have you had a wheeze in your chest on any occasion? Do you use medication for asthma? They were also asked whether they had consulted a doctor and/or been hospitalized because of symptoms, and whether they still had these problems. The women were examined in order to detect signs of asthma (wheezing). Women confirming having asthma or wheezing, or using medication for asthma, or where physical examination revealed wheezing were defined as having asthma/obstructive symptoms. Hypertension/high blood pressure was defined as using anti-hypertensive medication and/or systolic pressure ≥160 mmHg and/or diastolic blood pressure ≥95 mmHg at the time of examination.

In the follow-up study in 1992–1993, women reported whether they had asthma, whether they had had nervous symptoms and/or headache and/or abdominal symptoms (diarrhoea, nausea, constipation and/or abdominal pain) during the last 3 months, and whether they had seen a doctor because of any of these symptoms. They were interviewed by a physician with the following questions: Have you had a wheeze in your chest on any occasion? Do you use medication for asthma? The women were examined in order to detect signs of asthma (wheezing). The same determination for asthma/obstructive symptoms and hypertension/high blood pressure as in 1968–1969 was used.

Social and lifestyle-related variables

Women were classified according to whether they had grown up in the city or in the countryside, and according to whether they lived in a house or an apartment. Women reported if they had children or not. Women with only “basic” education (c. 6 years, depending on cohort) were classified as less educated. Women reported their own occupation and income and, if they were married, their husbands’ occupation and income. This information was transformed according to Carlsson's standard occupations grouping system [10]. For the purpose of this study, we combined group one (large-scale employers and officials of high or intermediate rank) and groups two and three (small-scale employers, lower-rank officials, and supervisors) into a “high and medium social class” group. Groups four and five (skilled and unskilled workers) were classified as a “lower social class” group.

Women were classified according to their smoking habits as smokers (women who smoked or had stopped smoking during the previous year), ex-smokers (women who had stopped smoking more than a year before examination) and non-smokers (women who never smoked). Women reported experience of suffering from stress by answering the following question: “Have you experienced stress once or for several periods during the last/5 year/s, or constant stress during the last/5 year/s? Stress is defined as having been irritable, tense, worried, frightened or anxious, or having suffered from insomnia related to worrying about work, health, family, conflicts with other people or for other reasons.”

Statistical methods

The chi-square test was used to test the hypothesis of no differences in prevalence of psychosomatic symptoms between the two groups (women with/without UPS in 1968–1969). The age-adjusted odds ratios (OR) and their confidence limits were calculated according to the Mantel-Haenszel procedure. Associations between sociodemographic variables, education, residence, stress, number of children, smoking and psychological symptoms, with UPS as the dependent variable, were tested in a logistic regression model both in 1968–1969 and 1992–1993.

Ethical approval and participant consent

The Ethics Committee of Göteborg University approved the study. All subjects in the 1992–1993 examinations gave informed consent, in accordance with the provisions of the Helsinki Declaration.

Conflict of interest

The authors have declared that no conflict of interest exists.

Results

Prevalence of unspecified psychological and psychosomatic symptoms

A total of 1462 women born in 1908, 1914, 1918, 1922 and 1930 were included in our analysis. In 1992–1993, 830 women participated. Table I shows the number of participants in 1968–1969 and 1992–1993 in the different age groups, and the frequency of reports of current or previous UPS in 1968–1969 and 1992–1993, as well as the women who reported UPS both in 1968–1969 and in 1992–1993.

Table I.  Year of birth and number of participants (n) and participation rate in the different age groups, and indication of unspecified psychological symptoms in 1968–1969 and 1992–1993, as well as women who indicated unspecified psychological symptoms both in 1968–1969 and 1992–1993.

1968–1969				1992–1993					1968–1969 and 1992–1993
Year of birth	Participants	Unspecified psychological symptoms		Participants	Unspecified psychological symptoms in 1968–1969		Unspecified psychological symptoms in 1992–1993		Unspecified psychological symptoms, 1968–1969 and 1992–1993
	n	n	(%)	n	n	(%)	n	(%)	n	(%)*
1908	81	29	(36)	19	7	(37)	3	(18)	1	(14)
1914	180	50	(28)	79	20	(25)	14	(19)	7	(35)
1918	398	95	(24)	220	46	(22)	39	(18)	22	(49)
1922	431	128	(28)	266	72	(25)	56	(21)	27	(38)
1930	372	112	(30)	252	78	(31)	53	(21)	31	(40)
Total	1462	414	(28)	836	223	(27)	165	(20)	88	(39)

* Percentage of women with unspecified psychological symptoms in 1968–1969.

Unspecified psychological symptoms and psychosomatic symptoms in 1968–1969

In 1968–1969, 24–36% in the different age groups (a total of 414 women) reported current or previous UPS. Table II shows the frequency of different psychosomatic symptoms (high blood pressure, asthma/obstructive symptoms, headache and abdominal symptoms) in women who stated that they had, had had, or did not have UPS in the 1968–1969 examination.

Table II.  Frequency of different psychosomatic symptoms (high blood pressure, asthma/obstructive symptoms, headache and abdominal symptoms) in the group of women with/without unspecified psychological symptoms in the 1968–1969 examination.

	High blood pressure		Asthma/obstructive symptoms		Headache		Abdominal symptoms
Psychological symptoms	Yes (%)	No (%)	Yes (%)	No (%)	Yes (%)	No (%)	Yes (%)	No (%)
Year of birth
1908	33.3	36.7	21.4	38.8	46.4	30.2	61.5	23.6
1914	32.1	27.0	33.3	25.6	36.0	21.9	39.2	19.8
1918	23.3	23.7	36.0	19.8	32.7	18.0	33.8	17.4
1922	24.0	30.1	33.6	28.3	40.5	22.8	39.6	24.2
1930	33.3	30.0	35.6	27.6	38.6	22.4	44.0	21.6
Total	27.8	28.3	34.4**	26.1	37.6***	21.7	40.0***	21.2

** p<0.01; *** p<0.001.

Unspecified psychological symptoms and psychosomatic symptoms in 1992–1993

In 1992–1993, the number of participants was 836, of whom between 18 and 21% in the different age groups (a total of 165 women) reported UPS (Table I). There were no statistically significant differences in prevalence between the different age cohorts. Of the participating women who had reported UPS in 1968–1969, only 39% reported UPS in 1992–1993 (Table I).

Associations between prevalence of unspecified psychological symptoms and psychosomatic symptoms in 1968–1969, as studied cross-sectionally

The prevalence of several of the psychosomatic symptoms differed significantly between women in 1968–1969 who had reported UPS and women who had not, when adjusted for age. Specifically, women with UPS had significantly higher prevalences of asthma/obstructive symptoms (OR 1.45, 95% CI 1.14–1.86), headache/migraine (OR 2.19, 95% CI 1.74–2.76) and abdominal symptoms (OR 2.53, 95% CI 2.00–3.18). With adjustments made for smoking, the women with UPS still showed significantly higher prevalences of asthma/obstructive symptoms. The prevalence of hypertension/high blood pressure, however, was not higher in the group of women with UPS.

The two groups of women, those with and those without UPS, were compared regarding residence, number of children, education, socioeconomic status, smoking and experience of suffering from stress. There was a statistically significant difference between the two groups regarding residence. Women who reported not having UPS lived in apartments to a lesser extent than women with UPS (OR 0.71, 95% CI 0.55–0.92).

Women with UPS were smokers (OR 1.63, CI 1.29–2.05) and experienced stress (OR 8.18, CI 6.20–10.80) to a higher extent than women without UPS in 1968–1969.

Associations between prevalence of unspecified psychological symptoms in 1968–1969 and psychosomatic symptoms after 24 years (longitudinal data)

The prevalence of some of the symptoms defined in our study as psychosomatic differed significantly between women who had reported UPS in 1968–1969 and women who had not. Specifically, women with UPS in 1968–1969 had a significantly higher prevalence of asthma/obstructive symptoms (OR 2.00, CI 1.34–2.98) and abdominal symptoms (OR 1.66, CI 1.16–2.40) in 1992–1993. Table III compares the frequency of psychosomatic symptoms in 1992–1993 in women with and without UPS in 1968–1969.

Table III.  Comparison of frequency of different psychosomatic symptoms in 1992–1993 in the groups of women with and without unspecified psychological symptoms in 1968–1969. The table presents individual cases linked from 1968–1969 to 1992–1993.

	High blood pressure, 1992–1993				Asthma/obstructive symptoms, 1992–1993				Headache, 1992–1993				Abdominal symptoms, 1992–1993
Year of birth	Psychological symptoms		No psychological symptoms		Psychological symptoms		No psychological symptoms		Psychological symptoms		No psychological symptoms		Psychological symptoms		No psychological symptoms
	Freq.	%	Freq.	%	Freq.	%	Freq.	%	Freq.	%	Freq.	%	Freq.	%	Freq.	%
1908	6/8	75	7/11	64	1/8	12	4/11	36	3/7	43	2/10	20	5/7	71	3/10	30
1914	14/20	70	45/59	76	4/20	20	9/59	15	8/20	40	10/55	18	10/20	50	21/55	38
1918	26/50	52	110/169	65	17/50	34	29/169	17	18/46	39	31/168	18	23/46	50	50/169	30
1922	39/74	32	102/191	53	17/74	23	22/191	17	16/72	22	43/190	23	28/72	39	61/190	32
1930	23/78	29	65/172	38	18/78	23	35/172	20	34/78	43	54/172	31	31/78	40	59/172	34
Total	108/230	47	329/602	54	57/230*	25	109/602	18	77/223	34	140/595	24	97/223**	44	194/596	33

* p<0.05; ** p<0.01.

Analysis of non-participants and deceased women in 24-year follow-up

The women who participated in the study in 1968–1969 but died during 1968–1992, and, as a consequence, did not participate in 1992–1993, had suffered from hypertension to a higher extent than the women who did not participate in 1992–1993 because of other reasons. There were no differences between the two groups of women regarding prevalence of asthma or headache in 1968–1969. The women who did not participate in 1992–1993 stated fewer abdominal symptoms in 1968–1969 than the participants in 1992–1993.

Concerning the social and lifestyle-related variables, there was a statistically significant difference between women who participated in 1992–1993 and women who did not participate. The women who did not participate in 1992–1993 were less educated and belonged to a lower social group than the women who participated. There were no differences between the women regarding having children, but women who did not participate in 1992–1993 were smokers to a higher degree than participating women.

Discussion

Our study shows that women with UPS in 1968–1969 simultaneously had significantly higher prevalences of asthma/obstructive symptoms, headache/migraine and abdominal symptoms, but not hypertension/high blood pressure. Among the women who had reported UPS in 1968–1969, a higher prevalence of asthma/obstructive symptoms and abdominal symptoms was found at the 24-year follow-up examination in 1992–1993, compared to women who did not report UPS in 1968–1969.

Clinical studies have shown a higher frequency of smoking among psychiatric patients than among control groups (52% vs 30%) [11]. Several recent studies have reported a close relationship between smoking and anxiety disorders (including panic disorders) [12–14]. Smoking can also cause higher blood pressure, but hypertension/high blood pressure was not significantly increased in the group of women with UPS.

The fact that there is an association between negative psychological symptoms and self-reported physical symptoms, but not necessarily objectively measured signs of physical disease [15], could also contribute to an increased tendency of reporting by women who also stated that they experienced UPS.

There was a difference between the two groups regarding residence. Women who had reported UPS lived in apartments more often than women without UPS. The association between morbidity and mortality indicators and low socioeconomic status has been observed for many centuries. Studies have shown that the lower the social class, the greater the amount and severity of perceived distress [14], [16]. Only a few follow-up studies have investigated the differences in psychosomatic health and socioeconomic status, and the association between them at different life stages. A Finnish study from 2005 showed an association between psychosomatic symptoms in adolescence and lower education in early adulthood [17].

Women with UPS also had more experience of suffering from stress than women without UPS. Stress could be assumed to be both a cause as well as a consequence of psychological symptoms. Does suffering from stress cause asthma and abdominal symptoms, or do the asthma and abdominal symptoms influence the mind and contribute to UPS?

Women with UPS in 1968–1969 seemed to have fewer psychosomatic symptoms in 1992–1993 than in 1968–1969. This could indicate that psychosomatic symptoms are self-limiting. Studies have also shown that the prevalence rate of psychological disorders in older primary-care patients is lower than in young patients [18].

There are not many studies concerning psychological symptoms and psychosomatic symptoms in a population over long periods of time. A prospective cohort study included 1537 children who were examined in 1942 with the aim of assessing the influence of childhood variables (physical and emotional) on later well-being in a group of rural Swiss (Emmental cohort). Emotional situation in the follow-up in 1995 was significantly correlated to psychological well-being at the beginning of the study. The somatic complaints in 1995 also correlated significantly to neurotic symptoms in childhood [19].

Population studies are valuable for obtaining important information about different diseases and symptoms, especially when the sample is followed for a long period of time. Valuable information is obtained from a cross-sectional study, but even more interesting results are to be expected from a longitudinal study. The advantages of this study are the cross-sectional and longitudinal design, following the women in the original study in 1968–1969 until the follow-up study in 1992–1993. Women were asked almost the same questions on both occasions. The development of the prevalence of psychological and psychosomatic symptoms could therefore be studied with a 24-year perspective.

There are some limitations to this study. The association between UPS and various psychosomatic symptoms was rather weak. This may partly be because we used the women's own description of psychological and psychosomatic symptoms rather than a more sophisticated medical instrument or doctors’ diagnoses. Secondly, the psychological and somatic symptoms were subjectively and individually reported. On the other hand, a symptom is a symptom and not a diagnosis or illness, and it is only the patient him/herself who can state whether or not he/she has the symptom.

Patients who experience symptoms presumably consult their doctors more often, producing the possibility of ascertainment bias. Illnesses such as hypertension and asthma could be diagnosed more often than in patients who do not consult as frequently, resulting in a higher probability of, for example, hypertension in patients experiencing psychosomatic symptoms. In spite of this, no connection between the prevalence of hypertension and UPS could be detected.

Conclusions

Women with UPS seemed to have a higher frequency of simultaneously psychosomatic symptoms than women without UPS. On the other hand, having UPS did not seem to generate further psychosomatic symptoms over time. Only 39% of women with UPS in 1968–1969 still had UPS in 1992–1993, and they differed from women without UPS in 1968–1969 concerning the prevalence of asthma/obstructive symptoms and abdominal symptoms, but not concerning headache and high blood pressure.

Acknowledgements

This study was funded by grants from the Swedish Research Council (345-2001-6652, 27X-04578, 2002-3724), the Bank of Sweden Tercentenary Foundation, and the Medical Faculty at the Sahlgrenska Academy at Göteborg University.