PERIMENSTRUAL SYMPTOMS (PS) Occur Immediately
Total Page:16
File Type:pdf, Size:1020Kb
Women's Health: women experience a marked increase in symptoms The Menstrual Cycle premenses. Cyclic variations in the prevalence of commonly Premenstrual Symptoms: cited perimenstrual symptoms were estimated from Another Look daily symptom recording. A community-based, multiethnic sample of 345 women recorded symptom severity from "not present" to "extreme"for 90 days. NANCY FUGATE WOODS, PhD, RN Maximum total reported symptom score occurred du- ring menses, not- during premenses. When individual Dr. Woods is Professor and Chair, Department of Parent and Child Nursing, University of Washington, Seattle, WA. This symptoms are considered, theprevalence ofthose rated research project was supported by a grant from the Division of as moderate to extreme during menses is less than 15 Nursing, USPHS No. NUO1054. This paper formed the basis of percent. her presentation at the National Conference on Women's Health, held in Bethesda, MD, June 17-18, 1986. A methodfor identifying symptom severity patterns Contributing to this paper were Martha Lentz, PhD, RN; Ellen Sullivan Mitchell, ARNP, PhD; Katherine Lee, PhD, RN; Diana throughout the menstrual cycle is described. Six Taylor, PhD candidate, RN; and Nitsa Allen-Barash, PhD candi- symptom severity patterns were identified. Only 13 date; all of the University of Washington School of Nursing. percent of the women exhibited a pattern of increased symptom severity in the premenses. Another 13 per- Synopsis ......................................................................... cent had a pattern of decreased symptom severity in the premenses. A collection of over 200 symptoms has been labeled premenstrual syndrome. Common belief is that most PERIMENSTRUAL SYMPTOMS (PS) occur immediately strual function is associated with ovarian activity, before and during menstruation and include irrita- investigators originally speculated that an imbalance bility, mood swings, depression, tension, anxiety, in the production of estrogen and progesterone was fatigue, cramps, backache, weight gain, painful or the factor precipitating PS (13). Although this tender breasts, and swelling. Although a collection hypothesis has been widely accepted, there have of nearly 200 symptoms has been labeled pre- been conflicting reports of ovarian steroid imbalance menstrual syndrome (PMS), these symptoms do not in the perimenstruum (14-18). Additional theories constitute a single syndrome, but occur in multiple have been proposed based on hypotheses of biologi- symptom configurations of negative affects, water cal excesses or deficiencies such as hyperaldostero- retention, and discomfort. Moreover, they are not nism, hyperprolactinemia, hypoglycemia, and vita- confined to the premenstruum, but occur during min and mineral deficiencies (19-27). Symptoms menses as well. Thus perimenstrual is a more apt associated with these excesses and deficiencies are descriptor (1-6). similar to those described as PMS. However, for One of the most striking characteristics of PS is its every etiological theory proposed, there is an equally variability, reflected both in the variety of symptoms convincing argument against it (17, 28-31). women experience and in the large variation in the Another biological line of investigation has ad- prevalence of these symptoms from one population dressed autonomic nervous system (ANS) altera- to another ( 7-12). Although 30-50 percent of U.S. tions that may account for PS. Some investigators women experience mild or moderate symptoms, only found no menstrual cycle phase differences in cate- 10-20 percent describe their symptoms as severe and cholamine excretion (epinephine and norepine- disabling (4). phine) in healthy women (32), or in skin con- Despite extensive research on the etiology and ductance, heart rate, or respiratory rate (33), while treatment of PS, the mechanisms producing them others found higher levels of ANS arousal in the remain unclear. Most etiological studies have fo- premenstrual phase as indicated by increased heart cused on either a single biological or behavioral rate, respiration rate, and body temperature (34). explanation for PS. Early work on PMS focused on Koeske (35) found a premenstrual increase in skin the biological etiology of symptoms. Because men- conductance levels which correlated with pre- 106 PUBLIC HEALTH REPORTS SUPPLEMENT menstrual moods. Coyne (36) found a statistically Our earlier prevalence estimates relied on retro- significant rise in premenstrual frontalis muscle ten- spective measures that required women to recall their sion levels that was exaggerated in women who symptom experiences over the most recent menstrual anticipated experiencing high levels of symptoms. period. Our comparisons of retrospective and pro- When symptomatic women were studied, cortisol spective measures revealed higher prevalence esti- levels were elevated in the luteal phase (37, 38), but mates for retrospective versus prospective reporting not above the normal range (39). Moreover, circa- (54). dian hormone secretory profiles did not differ in The purposes of this study were to women with PMS and asymptomatic women (40). Some investigators now suggest that PMS is a con- * Describe the prevalence of PS obtained from daily sequence of falling endorphin levels in the late luteal recordings in a multiethnic population. phase (40). * Describe the patterns of cycle phase change in Another line of investigation has addressed the symptom severity and their prevalence. influence of the social environment on symptom occurrence. Theories related to socialization include Method those linking individual differences in expectations about menstruation to symptoms (41). Socialization in a traditional feminine role has been linked to Sampling. The sampling framework employed in menstrual symptoms in some studies (42), but not in this study involved multiple steps. First, census block others (43, 44). The influence of social stressors on groups (fractions of census tracts) in which 40 symptoms has been demonstrated for major life percent or more of the population reported an in- events and daily stresses. Moreover, there is evi- come between $12,900 and $39,900 were identified dence for differential effects of stressors given the from the 1980 census data from King County, WA. nature of symptoms, with the major effect seen on Of those block groups meeting the initial income negative affects (44). criterion ( 901 ), in only 1 9 were 10 percent or more In controlled clinical trials of pharmacologic of the population, Black or Asian. Profiles of the agents such as progesterone (45), antiprostaglandins number of females between the ages of 15 and 59 (46-48), bromocriptine (22, 49), spironolactone years (these are the age bands employed in reporting (29, 50), pyridoxine (51, 52), magnesium (21), census data) and the educational status (completion tryptophan (51), and gonadotropin-releasing hor- of more than elementary school) of each ethnic mone (53), medical treatment has been no more group were generated for each of the 119 block effective than placebo therapy. Approximately 40 groups. The most suitable block groups from the percent of women respond favorably to inert placebo standpoint of age, ethnicity, and educational groups drugs. were then identified by the research team. The street Our early prevalence study (4) revealed that both segments of the selected block groups were then biological and social factors were associated with PS. identified and randomly ordered with a computer Parity and oral contraceptive use were associated program. The numbers of the street segments within with less severe cramping, whereas intrauterine de- block groups provided the link between this initial set vice use was associated with more severe cramping. of criteria and a city directory from which all poten- Women with long menstrual cycles and longer and tial participants' telephone numbers were obtained. heavier menstrual flows reported more severe PS, including water retention, cramps, and negative af- fects than did their counterparts with short cycles and Design. The hybrid design employed in this study less flow. Women who could accurately predict their involved cross-sectional, prospective, and retro- next menses also had more symptoms of weight gain spective elements. After households were screened and backache. In general, older women and em- to identify eligible women willing to participate in ployed women reported less severe symptoms than the study, participants completed an in-home inter- younger women. Black women reported less severe view and a 90-day health diary. After the women cramping and premenstrual negative feelings but completed their diaries, some were asked to partici- more weight gain, swelling, and headache than white pate in a food diary component of the study. Follow- women. Women exposed to stressful life circum- ing completion of the 90-day diary or the food diary, stances, including major life events and daily stresses or both, all women were interviewed by telephone. and who had few resources to deal with them (little formal education and low incomes), reported the Measures. Perception of symptoms referred to the most symptoms (4, 44). participants' daily rating of the presence and severity JULY-AUGUST 107 Table 1. Percentage of 345 women with perimenstrual symp- of symptoms from a 90-day health diary. The 57 toms and positive feelings reported In daily diaries as moder- items included in the health diary were generated ate or extreme, premenstrual and menstrual from