By Shea, J L
Key words: CHINESE WOMEN, CHINA, SYMPTOMS, MENOPAUSAL STATUS, AGE, ATTITUDES
ABSTRACT
Objectives The China Study of Midlife Women (CSMW) aimed to determine in mid-life Chinese women, first, the frequency of various symptoms often included in studies of menopause or the climacteric, second, attitudes toward menopause and aging, and, third, the relationship between symptoms and menopausal status, chronological age and attitudes.
Methods A questionnaire was administered via face-to-face interview to a general population sample of 399 Chinese women living in two communities in northern China.
Results The Chinese women displayed a low to moderate frequency of reporting 21 symptoms across vasomotor, vaginal, sleep-related, cognitive, emotional and somatic categories. Their attitudes toward menopause and aging tended to be more positive, neutral or ambivalent, as opposed to negative. There was a small, statistically significant association between six symptoms and menopausal status, four symptoms and chronological age, and 11 symptoms and negative attitudes toward menopause and aging.
Conclusions Across the measures utilized in this study, Chinese women’s symptom reporting is more strongly associated with their attitudes towards menopause and aging than their menopausal status or chronological age. More research is needed on cross-cultural, cross-ethnic and individual variation in women’s interpretations of statements often included in mid-life attitudinal scales.
INTRODUCTION
Scholars have long debated which symptoms are related to the menopause transition1, with some linking a broad range of physical and psychological symptoms and others a narrow set of physical symptoms. Several studies conducted in western countries have found that women’s attitudes toward menopause and aging may contribute as much or more to the production of various symptoms as either menopause or aging themselves do2-5. Studies of mid-life women in some non-Western societies have suggested that a more positive orientation toward menopause and aging in those societies helps to account for their relatively low symptom level6-10.
To explore these questions, this article draws on material from the China Study of Midlife Women (CSMW), a project the author conducted in mainland China over several years in the 1990s11,12. In relation to this sample, the analysis examines:
(1) The frequency of various symptoms commonly discussed in the research literature on menopause or the climacteric;
(2) Whether any of these symptoms demonstrate an association with menopausal status and/ or chronological age;
(3) The frequency of various attitudes toward menopause and aging; and
(4) The relationship between various symptoms and women’s attitudes toward menopause and aging.
METHODS
This research was conducted in a rural village and an urban neighborhood in the municipality of Beijing in 1994. These communities were designated by the Chinese government as respectively rural (nongcun) and urban (chengshi) based on their relative population densities, proximity to agricultural production and other factors. Each of these two sites was a middle-income community accessible for study due to positive grassroots rapport established over many years. Project approval was secured from the Institutional Review Boards of Harvard University and Peking Union Medical College Hospital, the Central Ministry of Public Health in China, and local leaders in the surveyed communities. Based on residency lists secured from the local urban and rural residency committees, a total sample was taken of mid-life women between the ages of 40 and 65 years in the village and a block of the neighborhood. At the time of the study, permission to do survey research was extremely difficult for foreign-led research projects such as this to attain, and, as a result, random samples of numerous communities were not possible. Informed consent was requested from each woman, and the participation rate was over 90%. A total of 420 women participated. In this analysis, data from 399 of these women are analyzed. The remaining 21 cases were set aside due to missing data or artificially induced menopause. None of the 399 women examined in this analysis had undergone surgical menopause.
Table 1 Sample characteristics of the Chinese women in the study, age 40-65 years
The sample was divided fairly evenly between the urban and rural sites and between women aged 40-49 and 50-65 years (Table 1). Based on standard definitions13,115 women were premenopausal, 98 perimenopausal, and 186 postmenopausal. Although more educated than the national average for women of their age, a broad range of educational levels was represented, with almost one-quarter having 3 years or less of schooling. A wide variety of occupations, from factory and service jobs to educational, managerial and technical work, was represented at both field sites, but the rural sample was distinguished by having a considerable proportion of women who were currently doing collective agricultural labor or raising pigs or chickens at home.
The questionnaire included a checklist asking women whether they had experienced various symptoms over the previous 2 weeks. Presentation of this checklist was not in any way linked with menopause or aging. The 21 checklist symptoms analyzed here (Table 2) were chosen for their association with biomedically oriented menopausal indexes13 and/or cross-cultural research projects on female mid-life9. The Chinese translations were developed through months of participant observation in Beijing, reading Chinese clinical and self-help literature, checking with local researchers, and pilot testing on local women14.
Table 2 Frequency of symptoms in previous 2 weeks in 399 mid- life Chinese women between the ages of 40 and 65 years
This part of the questionnaire addressing respondents’ attitudes toward menopause and aging was presented after and at a distance from the symptom checklist. Some items were derived from other menopause questionnaires1, and some from Chinese self-help books or everyday conversations with Chinese residents. Comprising 19 statements on the end of menstruation and 20 statements on the transition from middle to old age, they were introduced as ‘something that some people say’ (shuofa huo chuanshuo). Women were asked to give their opinion on each statement by answering ‘agree’, ‘disagree’, ‘it depends’, or ‘unsure’.
As many women were accustomed to stating their age by the lunar calendar and most were unfamiliar with formal definitions of menopausal status, questions allowing the researcher to calculate these in standard biomedical terms were included. Women were designated as premenopausal if they had menstruated during the past 3 months with regular periods over the past 12 months. Perimenopausal status was defined as menstruating during the past 12 months but not during the previous 3 months, or during the past 3 months but with increased irregularity. Postmenopausal status was defined as not menstruating within the previous 12 months. Women who had had a hysterectomy or oophorectomy, or who had undergone chemotherapy or radiation, were excluded from analysis.
The questionnaire was administered to each respondent in a face- to-face interview in women’s homes or an offset corner of a public space. Interviews were conducted in Mandarin Chinese by a team of eight local Chinese professional women following extensive training14.
Responses were coded numerically and entered into a database. Attitudinal items required special attention. After entry in raw form, attitudinal responses were coded on a scale of 1 to 4, with 4 representing the negative pole, and 1, the positive. Depending on the question, either agree or disagree was coded 4, with its opposite coded 1. ‘It depends’ was coded 3 as an ambivalent answer, and ‘unsure’ was coded 2 as a neutral response. These codes were multiplied by response frequency and added and then divided by four to create a negativity score for each item. A score close to 1.00 indicates a very positive attitude, near 4.00 a very negative one, and near 2.5 ambivalence or neutrality.
After coding, regression analysis was conducted to see if there was any relationship between checklist symptoms and either menopausal status, chronological age, the Menopause Attitudes score, or the Aging Attitudes score. Different techniques were used depending on the combination of continuous and categorical variables. Standard regression analysis was conducted when both outcome and predictor were continuous variables. Logistic regression was used when there was a dichotomous outcome and a continuous predictor. ANOVA-derived regression was used when both the dependent and independent variables were categorical.
RESULTS
Symptom frequency
Around one-half of the sample reported memory problems, feeling irritable and backache. Less than 10% reported night sweats, cold sweats and vaginal pain during sex (although three-quarters reported being sexually active12). A very low rate was found for all the vasomotor and vaginal symptoms, which were reported much less frequently than most sleep-related, cognitive, emotional, and somatic symptoms. The rate for hot flushes was slightly over 10% and, for vaginal dryness, under 15%. Difficulties sleeping were reported by about one-third. Within the cognitive and emotional categ\ories, around half of the sample reported memory problems and feeling irritable, but there were lower rates for difficulty concentrating, melancholy symptoms, and anxiety. Conceptually linked with the mid-life: transition in Japan and found in over one-half of mid-life Japanese women9, stiff shoulders was reported by less than 15% of the Chinese sample. (In interpreting the overall symptom frequencies given in Table 2, one needs to remember that over one- quarter of the sample was premenopausal and having regular menstrual periods (28.8%). Frequencies of six symptoms, including hot flushes, feeling tired, night sweats, insomnia, memory problems and anxiety, were significantly higher among periand postmenopausal women than in the overall sample including premenopausal women.)
Relationship of symptoms to menopausal status and age
Regression analysis showed that most of the 21 symptoms were not associated with menopausal status or chronological age (Table 3). However, night sweats, insomnia, memory problems, and anxiety were associated with both menopausal status and age, while hot flushes and feeling tired were related to menopausal status only. Although these were statistically significant relationships (p
Frequency of attitudes
On all 19 items on the Menopausal Attitudes Scale, the Chinese women gave more positive, neutral or ambivalent responses than negative ones. Reflecting this, their overall score on the Menopausal Attitudes Scale was 2.67. Six items exhibited a somewhat higher relative frequency of negative responses than typical for the rest of the scale; these items linked the end of menstruation with rapid aging, susceptibility to illness, senescence, feeling irritable and losing one’s temper. Comparing across menopausal status, there were relatively minor differences in the distribution of attitudes. The statement about the end of menstruation harming women’s health exhibited the largest difference, with premenopausal women giving the least positive response distribution. Across all items, the postmenopausal women had the most positive distribution of attitudes toward menopause (2.57) and the perimenopausal women the least (2.92), with the premenopausal women hovering in between (2.64).
With a score of 2.29 on the Aging Attitudes Scale, the Chinese women’s attitudes toward midlife aging were also mostly positive, neutral or ambivalent. Only one item out of 20 had a negative- leaning distribution; it was comprised of a well-known Chinese aphorism that, upon entering the mid-life transition, one lacks the energy and strength to do the things that one is interested in doing. In addition, six items with a positive distribution had more negative responses than typical. Three link middle age with rapid aging, susceptibility to illness, and regret over not realizing one’s aspirations. The others relate to losing one’s temper easily in the mid-life transition, being afraid of becoming a burden in old age, and younger generations not listening to their elders. The largest difference across age groups was displayed for a statement about women not noticing that they are aging if their lives are full of interesting activities, with women in their forties the least positive. Overall, there was a small difference by age in attitudes toward mid-life aging. Women in their fifties had the most positive attitudes toward aging (2.29) and women in their sixties the least positive (2.71), with the forty-somethings in between (2.45).
Table 3 Symptoms, menopausal status and age among 399 mid-life Chinese women between the ages of 40 and 65 years
On both scales, a large proportion of respondents replied either ‘it depends’ or ‘unsure’ instead of agreeing or disagreeing. In the interviews, many said that they did not know enough to state an opinion or that one could not generalize on these matters because every person or situation was different. Overall, the Chinese women’s score on the Menopause Attitudes Scale (2.67) was slightly less positive than on the Aging Attitudes Scale (2.29). This difference appears to be largely a function of menopause being seen as less salient in life than generic aging11, making for more unsure and ambivalent answers. In addition, the mixture of statements in the Aging Attitudes Scale may be more conducive to positive responses.
Relationship of symptoms to attitudes
Regression analysis showed some statistically significant associations (p
Table 4 Symptoms and attitudes among 399 mid-life Chinese women between the ages of 40 and 65 years
DISCUSSION
Frequencies of symptoms
Placing these symptom frequencies in crosscultural context, the hot flush rate among these Chinese women is considerably lower than the findings of many studies conducted in western countries; the rates of many other symptoms, however, are comparable4,15. A recent metaanalysis conducted by the National Institutes of Health summarized the sample frequencies for selected symptoms across numerous studies of mid-life women. They found a wide range of results, with anywhere from 14 to 80% for hot flushes, 4-39% for vaginal dryness, 16-60% for sleep disturbance, and 8-38% for mood symptoms16. This Chinese sample’s rate for hot flushes falls below the NIH range, but its rates are mid-range for vaginal dryness, sleep disturbance and most mood symptoms, except for irritability, which exceeds the NIH range. It is difficult to interpret these comparative results because of the tremendous methodological variation across studies comprising the NIH metaanalysis. Those studies differed in terms of the time frame specified for symptom recollection (e.g. previous 2 weeks/month/year, during menopause, ever before), the precise age span of the sample (e.g. age 45-55, 40- 60, 40-65 years), the stage(s) of the menopausal transition being examined, and whether surgically menopausal women were included. In this case, over one-quarter of the Chinese women were premenopausal and no surgically menopausal women were included, whereas, in the NIH meta-analysis, some studies did not have any premenopausal women and some included surgically menopausal women.
In this light, it must also be remembered that the 10.5% Chinese hot flush figure reported above is indexed to the previous 2 weeks. By contrast, the lifetime hot flush rate for the Chinese women in this present study was 32.4%, which is placed low-middle in the NIH range. This rate is similar to the lifetime rate found by Xu and colleagues17 in their 1990 community study of over 5000 mainland Chinese women also aged 40-65 years (36.8%). With respect to popular generalizations about Asian women, it is interesting to note that this Chinese rate is somewhat higher than the lifetime rate found among Japanese women (19.6%) by Lock18. Even when we limit the Chinese sample to the 45-55-year age range used in Lock’s Japan study, a higher relative Chinese lifetime hot flush rate remains (40.4%). This indicates that, while the short-term rate is similar for each sample (China 13.5%, Japan 12.3%), a different dynamic may operate over longer time frames.
Frequencies of attitudes
Overall, it was quite common for the Chinese women to express uncertainty concerning their attitudes toward menopause and women’s mid-life aging, probably due to little societal emphasis on the topic at the time of the study. Most either gave positive answers or said they didn’t know or ‘it depends’. The tendency among the Chinese women for attitudes toward menopause and aging to be positive, neutral or ambivalent is consistent with several other studies conducted in western and non-western countries9,19,20. Precise cross-cultural comparison of the results is difficult, however, because different studies have been based around varying attitudinal measures. Cross-cultural comparison of such attitudes is particularly challenging due to issues of translation, both in questionnaire design and in women’s interpretations of items.
The results on subgroup variation in Chinese women’s attitudes run in opposite directions with regard to Neugarten’s2 experiential hypothesis. The results of the present study support the notion that women’s views of menopause are more positive when personal experience is gained, but run counter to the idea that experience with generic aging mitigates negativity. These results underscore the importance of distinguishing menopause from aging in attitudinal measures; while overlap exists, there are important distinctions in women’s orientations toeach concept, a nuance often missed in the research literature.
Relationships between symptoms, menopausal status and age
In the recent NIH report16, the relationship between various symptoms and menopausal status was also examined. The NIH consensus was that most studies examined found that only hot flushes, night sweats, vaginal dryness and perhaps trouble sleeping are related to the menopausal transition. The finding of an association in the Chinese sample of menopausal status with hot flushes, night sweats and insomnia is consistent with the NIH analysis. The Chinese findings linking memory problems, anxiety and feeling tired with menopausal status, however, are not. This difference may be in part related to the fact that the NIH findings are a crossstudy generalization, de-emphasizing variation across populations.
As menopausal status and age naturally co-vary (Table 1), it is not surprising that there was considerable overlap in their symptom associations in the Chinese study. That said, there was a somewhat stronger relationship between the examined symptoms and menopausal status, as opposed to age. At the same time, however, none of the associations was large, perhaps because of the wide variety of factors that contribute to the production and expression of symptoms like these21.
Relationships between symptoms and attitudes
The Chinese results support the findings in other populations that women with negative attitudes toward menopause and aging tend to have higher levels of symptom reporting2-5,22,23. Furthermore, the Chinese study showed that negative attitudes accounted for almost twice as many symptoms as menopausal status and age did.
In some other studies, an association has been found between both menopausal status and women’s attitudes, at least for hot flushes4. In the Chinese study, hot flushes, night sweats, insomnia and anxiety were associated only with menopausal status and/or age and not with women’s attitudes. Lack of association between these symptoms and Chinese women’s attitudes may be related to how there was no strong conceptual association between these symptoms and female mid-life in China at the time. Instead, irritability was more strongly linked with notions of mid-life transition in China11,14. In this light, it is significant that irritability was associated only with women’s attitudes and not with menopausal status or age.
Limitations
This study is limited in several ways. First, as a cross- sectional study, it is uncertain whether relationships found here reflect a causal connection or just a correlation. Other factors co- varying with menopause and/or age may be the real mechanisms producing the experience and/or expression of these symptoms. Also, the direction of relationships is indeterminate. An association could mean that attitudes contribute to symptom production, but it could also mean that symptom experience shapes attitudes, or both.
Second, one of the attitudinal items on both scales – namely, the health-care statement – was interpreted by the Chinese respondents in two different ways, with a bearing on scoring. Observation revealed that most women took the statement as intended – that is, menopause is normal and natural, so women do not need extraordinary health-care measures. Some women, however, displayed a different interpretation. They retorted that older women have a right to health care and should not have to rely on taking care of themselves. These women did not necessarily see menopause as a time of sickness, but they felt that mid-life women deserved special treatment as much as anyone else. As a result, scoring this minority’s disagreement with this item as ‘negative’ is problematic. Similar issues with varied interpretation of attitudinal items may help to account for puzzling cross-ethnic results found with regard to Asian-American women in the US-based SWAN study24,25.
Third, these results cannot be assumed to be representative of all Chinese women. The sample size was modest, and the women studied were living in the capital region of the People’s Republic of China, an area considerably more developed economically than many other parts of China. In addition, sweeping social change has transformed China over the past decade since this study was conducted, and this cohort of women, born between 1928 and 1954, is quite different from later generations traversing mid-life. For example, mid-life women in China today tend to have had more lifetime exposure to formal educational opportunities, biomedical ideas and practices, and western media, diet and lifestyle.
Fourth, menopausal status was designated according to self- reported menstrual history, and hormonal assays were not taken in this study. It is unclear whether and how these women’s self- reports related to physiological changes in hormonal profile.
IMPLICATIONS AND DIRECTIONS FOR FUTURE RESEARCH
This study found that the mid-life Chinese women studied tended to have a lower prevalence of hot flushes, but a similar rate of reporting of many other symptoms, as compared to mid-life women in North America and many other western countries. The proclivity found among these Chinese women to eschew negative attitudes toward menopause and aging is consistent with several other studies conducted in both western and non-western countries, as is the link they demonstrated between negative attitudes and symptom reporting. Further research should investigate whether the low to moderate rate of symptom reporting and the relatively positive attitudes toward menopause and aging found in this study extend to current generations of mid-life women in the Beijing area and other regions of China. In addition, such research needs to examine whether there is a widespread tendency for negative attitudes toward menopause and aging among Chinese women to have a stronger association to symptom reporting than either menopausal status or chronological age do. In a broader context, more research is needed in a wide variety of settings world-wide to elucidate cross-cultural, ethnic and individual differences in women’s interpretations of various items commonly used in attitudinal scales on menopause and aging. Furthermore, additional information is needed on the degree to which various women differentiate their views of menopause from their views of aging. Such ethnographic knowledge is vital for understanding the local and individual meanings that may contribute to women’s experiences of mid-life and the production, expression and interpretation of symptoms therein.
ACKNOWLEDGEMENTS
Throughout the research, feedback was provided by James L. Watson, Arthur Kleinman, Rubie S. Watson, Michael Phillips and Xu Ling. Any errors are the author’s own.
Conflict of interest Nil.
Source of funding Financial support was provided by the Mellon Foundation, FLAS, CSCC, NSF, Chiang Chingkuo Foundation, NIMH, Cora DuBois Trust, Freeman Foundation, Lintilhac Foundation, Parimitas, and University of Vermont Dean’s Fund.
References
1. Lock M, Kaufert P. Menopause, local biologies, and cultures of aging. Am J Hum Biol 2001;13: 494-504
2. Neugarten BL. Women’s attitudes toward the menopause. In Neugarten BL, ed. Middle Age and Aging: A Reader in Social Psychology. Chicago: University of Chicago Press, 1968: 195-200
3. Hunter MS. Psychological and somatic experience of the menopause: a prospective study. Psychosom Med 1990;52:357-67
4. Avis N, McKinlay S. A longitudinal analysis of women’s attitudes toward the menopause: results from the Massachusetts Women’s Health Study. Maturitas 1991;13:65-79
5. Olofson ASB, Collins A. Psychosocial factors, attitude to menopause, and symptoms in Swedish perimenopausal women. Climacteric 2000;3:33-42
6. Bart, P. Depression in middle aged women. In Gornick V, Moran B, eds. Women in Sexist Society. New York: Basic Books, 1971:99-117
7. Flint MP. Sociology and anthropology of the menopause. In van Keep PA, ed. Female and Male Climacteric. Lancaster: MTP Press, 1978: 1-8
8. Kaufert P. Myth and the menopause. Social Health Illness 1982;4:41-66
9. Lock M. Encounters with Aging: Mythologies of Menopause in Japan and North America. Berkeley: University of California Press, 1993
10. Berger GE. Menopause and Culture. London: Pluto Press, 1999
11. Shea J. Revolutionary Women At Middle Age: An Ethnographic Survey of Menopause and Midlife Aging in Beijing, China. PhD dissertation, Cambridge, Massachusetts: Department of Anthropology, Harvard University, 1998
12. Shea J. Sexual ‘liberation’ and the older woman in mainland China. Modern China 2005;31: 115-47
13. Kaufert P, Syrotuik J. Symptom reporting at the menopause. Soc Sci Med 1981;151:173-84
14. Shea J. Cross-cultural comparison of women’s midlife symptom reporting: a China study. Cult Med Psych 2006; in press
15. Avis N, Kaufert P, Lock M, McKinlay S, Vass K. The evolution of menopausal symptoms. Bailliere’s Clin Endocrin Metabol 1993;7:17- 32
16. NIH Conference statement: management of menopause-related symptoms. NIH state-of-the-science conference on management of menopause-related symptoms, Bethesda: National Institutes of Public Health, March 21-23, 2005
17. Xu L, Zhao X, Ge QS. Epidemiology of perimenopause. Shengzhi Yixue Zazhi (J Reprod Med) 1993;2:23-7
18. Lock M. Contested meanings of the menopause. Lancet 1991;337:1270-2
19. Kaufert P, Gilbert P. Women, menopause, and medicalization. Cult Med Psych 1986;10:7-21
20. Woods NF, Mitchell ES. Symptom experiences of midlife women: observations from the Seattle Midlife Women’s Health Study. Maturitas 1996; 25:1-10
21. Martinez-Hernaez A. What’s Behind the Symptom? On Psychiatric Observation and Anthropological Understanding. Amsterdam: Harwood Academic, 2000
22. Avis, NE, Crawford, SL, McKinlay SM. Psychosocial, behavioral, and health factors related to menopause symptomatology. Women’s Health 1997;3:103-20
23. Perz JM. Psychological and social concomitants of the female menopause\: a longitudinal study. Maturitas 1997;27:82
24. Sommer B, Avis N, Meyer P, et al. Attitudes toward menopause and aging across ethnic/ racial groups. Psychosom Med 1999;61:868- 75
25. Avis N, Stellato R, Crawford S, et al. Is there a menopausal syndrome? Menopausal status and symptoms across racial/ethnic groups. Soc Sci Med 2001;52:345-56
26. StataCorp, Stata Reference, A-F, College Station, Texas: Stata Press, 1997
J. L. Shea
Department of Anthropology, University of Vermont, Burlington, Vermont, USA
Correspondence: Professor J. L. Shea, Department of Anthropology, 515 Williams Hall, University of Vermont, Burlington, Vermont 05405, USA
Received 8-07-05
Revised 14-11-05
Accepted 18-11-05
ORIGINAL ARTICLE
2006 International Menopause Society
DOI: 10.1080/13697130500499914
Copyright CRC Press Feb 2006
Comments