Breakage and slippage of condoms in family planning clients.

Information was obtained from 540 clients on the 3,754 occasions when condoms were used over a one-month period. One or more major problems, breakage, slippage, leakage or a combination of these, occurred in 410 (10.9%) events, involving 217 (40.2%) clients. Breakage occurred in 209 (5.6%) events and slippage in 243 (6.5%). Leakage was almost always associated with breakage or slippage. There was a small group of accident prone clients. Breakage was more common in younger and inexperienced clients and in those who had previously experienced breakage. There were some differences in breakage rates between brands of condoms. Poor fitting condoms and non-spermicidal condoms were associated with more breaks. Other contributing factors included vigorous sex, dryness and tearing with fingernails. Additional lubrication did not protect from breaks and saliva was associated with more breaks than expected. Oil-based lubricants were associated with less breaks than expected. On many occasions (66.0%), clients were aware of the break before ejaculation occurred. Slippage was more common in the inexperienced and in those who had previously experienced slippage. There was no association between brand of condom and slippage but poor fitting condoms were associated with more slips. The most common reason for slippage was the condom being left on too long. Additional lubrication did not affect slippage.

PIP: The authors collected information from 540 family planning clinic attendees on 3754 occasions when condoms were used over a one-month period to assess the incidence of breakage, slippage, leakage, or a combination of these events. 117 participants were male, 423 female, 23.1% under age 20, and only 28.2% over age 30. Problems occurred in 410 (10.9%) events, involving 217 (40.2%) clients. Breakage specifically occurred in 209 (5.6%) events and slippage in 243 (6.5%). Clients under age 30 were responsible for 87.1% of breaks. Of particular interest, oral and anal sex conferred 4.2 times the risk of breakage as vaginal sex, and 2.6 times the risk of slippage. Moreover, a small group of 16 people experienced 3 or more breaks during the month, 10 also experienced slippage. This 3% of 540 participants accounted for 71 breaks, or 34% of the 209 events. They tended to be young and less experienced by years of condom use. 11 mentioned vigorous sex, 9 mentioned dryness, and 9 mentioned condoms being too small as potential factors for condom problems encountered. The authors found complaints about condoms being either too small or too large predictive of an higher incidence of breakage and slippage. Differences were observed in breakage rates between condom brands. Clients in 66.0% of breakages were aware of the occurrence of breakage before ejaculation occurred. Additional lubrication did not protect from breaks and saliva was associated with more breaks than expected, while oil-based lubricants were associated with less breaks than expected.

Pill method failures.

This is a study over a four year period from November 1981 to December 1985 documenting 163 cases of pill method failures in reliable pill takers. In over one third of these cases (36%) there were no predisposing factors. Significant factors which were associated with failure were diarrhoea and/or vomiting in 35%, and breakthrough bleeding on the combined pill in 21%. While there is controversy in the literature about the role of antibiotics in causing pill failure, this study found 23% of failures associated with antibiotic use. Two failures occurred on anticonvulsant medication. Recommendations are made for improved instructions to patients.

PIP: Over the November 1981 to December 1985 period, a questionnaire was completed by the counselor on 163 patients attending the Parkview clinic (Wellington, New Zealand) for termination of pregnancy due to failure of oral contraception (OC). The reasons for selecting reliable pill takers for special study were: to obtain data on the characteristics of this group; to confirm to what extent previously recognized factors in OC method failure are operating in New Zealand; to determine the numbers who failed on antibiotic medication; and to explore other factors not previously studied such as weight, smoking, stress, and amenorrhea; and, on the basis of the findings, to make practical recommendations for improved instructions in the use of OC. Patients were excluded if there was any question regarding their reliability in OC use. The questionnaire was completed after the abortion had been performed to minimize the effect of withholding information in order to present a better case for termination. 41 patients (25%) were aged 15-19 years, 60 patients (37%) were aged 20-24 years, 40 patients (25%) were aged 25-29 years, and 20 (12%) were aged 30 years and older. 91 patients were under 70 kilograms, and 17 patients were 70 kilograms or more. 82 patients (50%) were nonsmokers; 68 patients smoked more than 10 cigarettes a day (42%), and 13 patients smoked less than 10 cigarettes a day (8%). In 27 patients (7%), there was a history of previous pill failure and in 6 patients there was a previous failure while using an IUD. In 8 of the 27 patients who experienced a 2nd pill failure, there was no obvious predisposing factor on this occasion. In 11 cases (7%) failure occurred in the 1st month of use. In 51 cases (31%) it occurred between 1-6 months. In 35 cases (21%) it occurred between 6-12 months. In 50 cases (31%) it occurred between 1-5 years. In 16 cases (10%) patients had been on the for more than 5 years. In the 11 cases where failure occurred in the 1st month of use, 7 patients had been using another brand in the month before, and this was not necessarily a higher dose pill. In 3 cases they were starting the pill after abstinence and in 1 case after the use of a barrier method. In 6 cases predisposing factors such as diarrhea and/or vomiting, antibiotic use, or breakthrough bleeding were present, but in the remaining 5 cases there were no known predisposing factors. Excluding patients who had failed on a progestogen-only pill when irregular menses may occur, there were 29 patients who gave a history of breakthrough bleeding on the combined pill (21%). This associated with other predisposing factors in 18 cases (13). Vomiting only was associated with 14 failures (9%); diarrhea only was associated with 23 failures (14%). Diarrhea and vomiting was associated with 19 failures (12%). In 37 cases (23%), failure was associated with the use of antibiotics. 2 failures occurred on anticonvulsant medication. Recommendations are made for improved instructions to patients.

The contraceptive practice of one thousand abortion patients.

This is a report on the contraceptive practice of 1000 cases who attended the Parkview clinic for termination of pregnancy. Fifty percent were not using any method of contraception at the time of conception. The principal reasons for non-use were dissatisfaction with previous methods and risk taking behaviour. Fifty percent became pregnant while using contraception, and in 20 percent methods which are generally regarded as reliable had been used. Fourteen percent were using the pill at the time of failure and this group is studied in greater detail. Improvements in contraceptive practice are suggested.

PIP: This report on the contraceptive practice of 996 women who attended the Parkview Clinic of the Wellington Hospital in Wellington, New Zealand for abortion aims to identify some of the reasons for contraceptive failure. Past and future contraceptive practice was fully discussed during the interview with counselors and certifying consultants. 26 patients were under age 16, 55 were aged 16 years, 214 were between 17-19 years, 500 were aged 20-29 years, 185 were aged 30-39 years, and 16 were 40-44 years. There was a significant overrepresentation of "other" Polynesians, especially Samoans. Half the patients (500) were not using any contraceptive method at the time that conception occurred. Of these, 98 had never used any method, and 402 had used methods at other times but not on the occasion believed to result in conception. Of young women aged 16 and under, 40.7% had never used any contraceptive method, 24.7% had used a method previously but not on the occasion resulting in the pregnancy. This compares with 56.4% of the age group 17-19 years and 45.7% of the aged group 20 and over. More women in the older age groups had tried contraceptive methods, but there was a high incidence of dissatisfaction with previously used methods. In sum, 50% of the women were not using any contraceptive method at the time of conception, 30% became pregnant while using those methods usually described as less reliable, and 20% while using reliable methods. Comments on method failures are provided. 1 failed vasectomy was due to recanalization 10 years after the operation, and the other was due to recanalization within the 1st year but after initial sperm clearance. 1 case of failed female sterilization was due to an unrecognized conception at the time of the operation. Details of the others which occurred at 3 months, 14 months, and 3 years after the operation were not recorded. There were no method failures with Depo-Provera, but 6 patients became pregnant when they failed to renew their injection at the 12-week interval. All types of IUDs had failed with the most common failures being the Copper 7 and the Multiload. Regarding the condom, no attempt was made to determine how many were due to patient failure and how many due to failure of the method with conscientious use. This was also the case regarding the diaphragm which was always used with a spermicidal preparation. Various methods of natural family planning were used. Withdrawal was often used in conjunction with other methods, and in a few cases it has been used alone and reliably for years before failiing. Misleading advertising was responsible for women relying on spermicides alone. The 140 oral contraceptive (OC) failures were examined in detail. 84 or 60% were considered to be patients failures and 56 or 40% method failures.

The contraceptive practice of abortion patients.

An analysis is made of the contraceptive practice of 100 women prior to abortion. Half the women were not using any method during the month they became pregnant. Problems with the pill, risk-taking behaviour and fears of infertility, were the commonest reasons for non-use. In the 50 percent who were using a method, failure was most commonly due to human error and to the use of less reliable methods. Psychological and social factors were seen to be important determinants of behaviour. This study emphasises the need for remedial action.

PIP: A survey was conducted among 100 women who attended a New Zealand family planning clinic with an unplanned and unwanted pregnancy and who proceeded to termination of the pregnancy. The survey assessed prior contraceptive usage. 1/2 of the women had been using no contraception during the month in which they became pregnant. Problems with the pill, risk-taking behavior, and fears regarding possible infertility were the reasons most commonly cited for nonpractice of contraception. Of the 50% who were using 1 or another method, the pregnancy was due mainly to use of less reliable methods or to human error. Social and psychological factors were shown to be important behavior determinants. The medical practitioner has an important role to play in influencing contraceptive behavior.

The use of behavioral methods of contraception in women seeking abortion.

AIMS: To determine the extent to which behavioural methods, used alone or with other methods of contraception, contribute to contraceptive failure in women seeking termination of pregnancy. METHOD: A clinical audit was conducted of the use of behavioural methods of contraception, in 1342 women attending the Parkview Clinic for termination of pregnancy, over a four year period 1992-6. The information was obtained through standard interview and patient records. RESULTS: The women seen were representative of New Zealand women undergoing termination of pregnancy. Approximately one third of women (34.9%) used one or more behavioural methods in the month prior to conception. The two most common methods were periodic abstinence and coitus interruptus, both used by approximately 17%. Other methods used less frequently and not solely for contraceptive purposes, were cleansing of the vagina and breastfeeding. Pacific Island and Asian women were more likely to use behavioural methods than European and Maori women. Many women (45.6%) using behavioural methods also used other methods of contraception, especially the condom. CONCLUSIONS: Behavioural methods were found more commonly than previously reported. Greater understanding of the use of behavioural methods will assist in the implementation of preventive programmes to reduce the number of terminations.

Condom failures in women presenting for abortion.

AIMS: To document the main reasons for condom failure in women presenting for first trimester termination of pregnancy. METHODS: From 1990-97 an audit was carried out, on the 3283 cases personally operated on by the author, using the information routinely obtained during pre-operative counselling. RESULTS: Two sets of figures were obtained. The first (minimum figures) were restricted to those who used condoms on every occasion of coitus (746 or 22.7% of all women seen). The second (maximum figures) included all of these plus those who did not use condoms on every occasion or who used condoms in conjunction with another method (1494 or 45.5% of all women seen). In this expanded group, there was no obvious cause for the failure in 446 (29.9% of condom users). The main reason for failure was not using condoms every time (736 or 49.3% of condom failures). Leakage of semen occurred in 321 cases (21.5% of condom failures) and 104 of these had used emergency contraception which had also failed (7.0% of condom failures). Condoms failed more often in women under 25 years of age. CONCLUSIONS: There is a need for better education on correct condom use and improved availability of emergency contraception. There is also a need for greater standardisation in reporting contraceptive failure.

Pill method failures in women seeking abortion: fourteen years experience.

AIMS: To document predisposing factors in 769 women who had inadvertently conceived while taking oral contraceptive pills reliably. METHODS: Over a 14-year period, 1981-1995, two questionnaires were administered, one for women using the combined oral contraceptive pill (COC) and one for women using the progestogen only pill (POP). RESULTS: More than one predisposing factor was usually present, the most common being diarrhoea and/or vomiting in 39.0%. Antibiotics had been prescribed in 20.5%. COC failure was associated with a high incidence of menstrual disturbance (37.1%). A history of a previous pill failure was obtained in 27.8%. Smoking was a risk factor for women on the COC. Only 6.0% became pregnant during the first month of pill use. As expected, POP users experienced a higher number of pill failures in proportion to the market share, but excess weight and time of taking did not appear to be risk factors. Most COCs failed according to the numbers expected from the market share. CONCLUSIONS: Many pill failures could be averted by better information about known or suspected risk factors, especially gastrointestinal upset and interacting medications. Pill failure is one more reason to discourage smoking in COC users.

PIP: Predisposing factors to oral contraceptive (OC) failure in reliable pill takers were investigated in a survey of 769 women who presented to Wellington (New Zealand) Hospital in 1981-95 for induced abortion. More than one predisposing factor was found in the majority of cases of pill failure. The highest percentage (37%) of pill failures occurred in women 20-24 years of age. The most common risk factor was diarrhea and/or vomiting, present around the time of conception in 39% of women. Another 21% took antibiotics during this period. Menstrual disturbance prior to the pregnancy occurred in 37.1%--a rate higher than that expected from studies of pill use. A history of previous pill failure was present in 27.8%. Smoking was a risk factor in users of combined OCs. 33.7% of women using the progestogen-only pill and 34.2% of combined OC users were under severe stress at the time pregnancy occurred. As expected, progestogen-only pill users experienced a higher rate of failure relative to the market share, but excess weight and the time of pill taking were not risk factors. These findings suggest a need for careful counseling about the potential of gastrointestinal illness and interacting medications to compromise pill effectiveness.

Sex information and attitudes in young people.

A questionnaire survey was conducted in the Wellington area among 696 senior secondary school pupils. The first part of the questionnaire studied aspects of sexual knowledge and the sources of information. The general level of knowledge was good at an elementary level but less than adequate when more detailed knowledge was considered. The role of peers, parents and schools was complementary. In the second part of the questionnaire a standard attitude test was administered. A majority of young people in this survey supported a standard of permissiveness with affection and did not support a double standard giving greater rights to males. Boys were more permissive than girls.

Pregnancies in reliable pill takers.

This study is an extension of a previous study on oral contraceptive pill method failure in reliable pill takers. It documents 137 cases which presented during the 3 year period from December 1985 to December 1988. The most important new finding is that smoking is a significant risk factor in failure on the combined pill (67% were smokers, n = 118), but not on the progestagen-only pill, although the numbers in the latter group were small (n = 19). Diarrhoea and/or vomiting (46%) and drug interactions (33%) are confirmed as important predisposing factors. Systemic illness and/or severe psychological stress may also affect pill utilisation and such factors were found in 47%. Menstrual disturbances were found in 32% of women on the combined pill and 68% of these were smokers. Multiple factors were present in most cases (79%). Repeat failures occurred in 25%.

PIP: A questionnaire was administered to the 137 women who presented to Wellington Hospital from December 1985-December 1988 for termination of pregnancy due to method failure while using oral contraceptives (OCs). Specifically investigated was the presence in these women of risk factors such as menstrual disturbances, cigarette smoking, diarrhea, vomiting, and interactions with other drugs. 118 of the pregnancies involved users of combined OCs; the remaining 19 women were using a progestogen-only pill. As in previous studies, diarrhea and/or vomiting emerged as the most significant factor in OC failure (46% of the 137 cases in this series). In an additional 34% of cases, OC failure was linked to the intake of various other medications (e.g., antibiotics) believed to interfere with OC utilization. 32% of the combined OC users with an unintended pregnancy reported menstrual disturbances such as irregular bleeding, breakthrough bleeding, or amenorrhea. Of interest was the finding that 68% of the subjects with menstrual disturbances were smokers. In addition, 67% of the women in the combined OC group who experienced method failure were cigarette smokers, a figure that is 20% than would be expected based on an analysis of census data on female smoking. Among users of progestogen-only Ocs, there were no significant association with menstrual disturbances and/or smoking with method failure; however, the sample size was small (19 women). Systemic illnesses and/or psychological stress were present in 47% of the women in this series. Multiple risk factors were present in 79% of the cases analyzed. 25% of subjects had experienced previous method failure with OCs.

Trends in the contraceptive practices of women seeking abortions in the 1980s.

The demographic features and contraceptive practices of 1000 women attending Parkview clinic of Wellington Hospital for termination of pregnancy were studied over an eight month period in 1988-9. Comparisons were made with a previous study at the same clinic in 1980-1. The overall abortion rate has increased from 6.8/1000 women in the Wellington statistical area in 1981 to 9.8 in 1989. The proportion of Pacific Island and Asian women presenting for abortions is high and has increased disproportionately between 1981 and 1989. The abortion rate has also increased in lower socioeconomic groups in 1989. The proportion of women using contraception at the time of conception increased from 50% in 1981 to 68.5% in 1989. The methods used by women presenting for abortion have changed significantly. There has been an increase in the proportion of women using condoms (from 13.3% to 36.2%) and the oral contraceptive pill from (14% to 21.4%).

PIP: Researchers analyzed data from 1000 abortion patients at the Parkview clinic at Wellington Hospital in New Zealand in 1988-1989 to compare them with those of 1000 patients in a comparable survey at the same clinic in 1980-1981. The overall abortion rate rose from 6.8-9.8, even though contraceptive nonuse by abortion patients consistently fell in the 1980s (50% in 1981; 38% in 1983; 31% in 1989). Age specific abortion rates were statistically different (p=.01). Between 1981-1989, the abortion rate fell for women =or- 16 years and increased in the other age groups. Even though most abortion patients were of European descent (64.8%), they were underrepresented when compared to their proportion in the local population (85.7%). Conversely, the women of Pacific Island descent and those of Asian descent were overrepresented. Indeed 13.1% of abortion patients were of Pacific Island descent while they represented only 3.1% of the population and 9.4% of Asians while they made up only 2.5% of the population. In 1989, 69% of abortion patients experienced contraception failure. 31% did not use contraception but 72% of them had used it in the past. In both studies, condoms and oral contraceptives (OCs) were the leading contraceptives. The most significant change in contraceptive behavior of abortion patients in the 1980s was the increase in condom use (13.3-36.2%; p=.0001). This may be attributed to the increased publicity about safe sex and condom use beginning in 1985. Nevertheless many used them incorrectly or inconsistently. OC use increased from 14-21.4%. Contraceptive behavior varied with ethnicity in 1989 (p=.001). For instance, 60.3% of Pacific Islanders did not use contraception at the time of the survey and 44% of them never used one. European and Asian women tended to use condoms while Maori women preferred OCs. Women in the lowers socioeconomic group were more likely to not use a contraceptive compared to other groups (p=.05).

"Precordial catch": a benign syndrome of chest pain in young persons.

A common syndrome of brief, sharp, severe precordial pain in 45 healthy young persons is described. The pain occurs at variable intervals, out of the blue. It typically occurs during rest or very mild activity and never on exertion. The onset is usually in adolescence. Most patients were of light or medium build, were nonsmokers and essentially healthy. It affects males and females. Recognition of the syndrome of precordial catch and reassurance of its benign nature will allay anxiety.

Postcoital contraception: a family planning study.

The New Zealand Family Planning Association undertook a prospective study of Yuzpe's postcoital method of contraception (0.1 mg ethinyloestradiol and 1 mg d1-norgestrel taken within 72 hours of unprotected intercourse and repeated 12 hours later). The study also used pills containing levonorgestrel. Both pill formulations were equally effective. All participants were drawn from six family planning branches throughout the country. The study covered a period of one year. There were 909 participants with 8% lost to follow up. Strict criteria excluded women on medication or hormone therapy. The ages ranged from 11 to 43 years with 92% aged 19 years and under. Results revealed an overall failure rate of 2.3% and a significantly higher failure rate (4.49%) if the method was taken after 48 hours of unprotected intercourse and a significantly lower failure rate (1.22%) if taken before 12 hours of unprotected intercourse. Vomiting occurred in 17% but it did not affect the failure rate.

A trial of two methods of taking cervical smears: the Aylesbury spatula plus cytology brush compared to the Cervex broom.

AIMS: To compare the adequacy of cervical smear taking, using the Aylesbury spatula plus cytology brush with the Cervex broom. METHODS: Two cervical smears were taken at the same visit, the women acting as their own controls. In Group 1 comprising 81 women, the first smear was taken using the Cervex broom. In Group 2 comprising 97 women, the first smear was taken using the Aylesbury spatula followed by the cytology brush. The tips of the samplers were sent to the laboratory in a cytology container with 30% ethyl alcohol in saline, for analysis of residual cells. RESULTS: It took experienced staff two or three smears before they were adept at taking an adequate Cervex broom sample as defined by the presence of endocervical cells. Both techniques were equally good at detecting significant abnormalities. When the Cervex broom was used first there were more smears with no blood present but when bleeding occurred there was no significant difference between the two groups. Both techniques were well accepted by the women. CONCLUSIONS: Although it is more expensive than the Aylesbury spatula and the cytology brush the Cervex broom has the advantage of allowing a simple one step procedure, thereby reducing the potential for air drying.

The prevalence of premenstrual symptoms in Wellington women.

A survey of 1826 women in the Wellington region was carried out. Participants were asked about their general and gynaecological health and their menstrual, contraceptive and obstetric history. Detailed questions were asked about the menstruating women's last menstrual cycle. Subjects were premenopausal women attending a random sample of general practice surgeries for any reason. The achieved sample is not intended as a cross-section of the Wellington community, but is nevertheless reasonably representative of the general adult female population aged 20 to 45. Most women volunteered one or more premenstrual symptoms; and about half of the women said these symptoms were such that they affected their lives.

Treatment of premenstrual symptoms in Wellington women.

A survey of 1826 women in the Wellington region was carried out. Participants were asked about their general and gynaecological health, as well as detailed questions about their last menstrual cycle. The majority (1456) had had a menstrual period within the last month or so. Eighty five percent of these women noted premenstrual symptoms of some kind, and were asked about a variety of self-help measures, and medical help, for these, and whether the advice and/or treatment was in fact helpful. Nine hundred and ninety women had tried self-help while four hundred and sixteen had sought medical help. The most commonly tried self-help measures were exercise, rest and vitamin B6. Half the women had tried each of these. Overall, there was a marked placebo response, but exercise, rest and keeping a written diary of symptoms were all helpful in over eighty percent of those who tried them. Doctors offered a wide range of treatments, including vitamin B6, diuretics, oral contraceptives and mefanamic acid, but the effect of these was difficult to evaluate further. When the sample was subdivided into clusters of women who shared similar symptoms, significant differences in the effectiveness of different self help measures emerged. Four different premenstrual syndromes are suggested: PMS-breast, PMS-bloat, PMS-irritable and intolerant, and PMS-various.

Vaginal dinoprostone versus oral misoprostol for predilatation of the cervix in first trimester surgical abortion.

Prostaglandins are effective in predilatation of the cervix prior to first trimester surgical termination of pregnancy under local analgesia. Arandomized open comparative trial was devised to compare the effectiveness and acceptability of vaginal dinoprostone with oral misoprostol. Two groups were randomized to control for age, parity and ethnicity. The operation was easier and less painful in older, parous, and Polynesian women. Both methods were effective with respect to ease of dilatation. Both were acceptable and equal with respect to the level of pain experienced by the woman during the operation. Vaginal dinoprostone is more gradual in its action, but it is more expensive, has to be refrigerated and self-insertion may sometimes cause problems. Oral misoprostol is considerably cheaper and does not require refrigeration, but it was associated with more preoperative nausea, cramping and occasional heavy bleeding.

PIP: The effectiveness and side effects of vaginal dinoprostone and oral misoprostol for cervical predilatation in first-trimester surgical abortion were compared in a study conducted at a New Zealand abortion clinic. Over a 3-month period in 1995, abortion patients at Parkview Clinic (Wellington South, New Zealand) were randomized to produce two groups matched for age, parity, and ethnicity. The first group (n = 153) received a vaginal suppository containing 3 mg of dinoprostone, which they inserted themselves at least 12 hours in advance of surgery; women in the second group (n = 160) were given 400 mcg of oral misoprostol at the clinic at least 1 hour in advance of abortion. Misoprostol was associated with more cramping, preoperative vaginal bleeding, and nausea than dinoprostone. Pain ratings did not differ significantly between groups. Overall, dilatation was slightly easier with misoprostol than with dinoprostone. The operation was also easier and less painful in older, parous, and Polynesian women. In the 54 cases where dilatation was rated as difficult, 48 women (88.9%) were nulliparous. Misoprostol is less expensive than dinoprostone and does not require refrigeration. As a result of these findings, Parkview Clinic has decided to use misoprostol for cervical dilatation.

Premenstrual syndromes defined by symptom-sets.

An analysis is made of the pattern of presenting premenstrual symptoms in randomly selected general practice patients from the Wellington region, New Zealand. Participants, 1826 healthy women 16-54 years old whose characteristics were reasonably representative of the adult female population, were asked about their general, obstetrical and gynaecological health. For the 1456 women who had menstruated within the last month or so, detailed questions were asked about the last menstrual cycle. Each woman was assigned to one of seven premenstrual symptom sets. Three groups had 'pure' symptoms, ie a predominant single symptom (breast tenderness, bloating or irritability). Three groups had 'mixed' symptom-sets. The largest of the 'mixed' groups was formed by the women who reported breast tenderness, bloating and irritability together with tension and depression. Women in this group were most likely to rate their symptomatology as severe. The last group contains a large number of women with miscellaneous symptoms. Characteristics of these groups are outlined. The study highlights the importance of distinguishing among premenstrual syndromes as this can foster more effective clinical management.