Contraception
46:359-367,
1992
USE OF A MONOPHASIC, LOW-DOSE ORAL CONTRACEPTIVE IN RELATION TO MENTAL FUNCTIONING J.B. Deijerl*, K.J. Duyn **, W.A. Jamen* and J.W. Klitsie** * Dept. of Psychophysiology, Vrije Universiteit, De Boelelaan 1111, 1081 HV Amsterdam, The Netherlands ** Wyeth Laboratoria B.V., Hoofddorp, The Netherlands
ABSTRACT The objective of the study was to evaluate the effect of Minulet, a new low-dose oral contraceptive on mood in two groups and to compare the effect with a control group of women not taking oral contraceptives (OC). The women participating were between 16 and 45 years of age. They completed the Amsterdam Mood Questionnaire (AMQ) and the Sickness Impact Profile (SIP) three times. They were filled in before treatment started, after taking Minulet for one month and then again after three months. The questionnaires were filled in by a group of 200 women who had not taken OC before (starters), and by a second group of 370 women who were already taking OC (switchers). A group of 140 women who did not use any OC during the study served as a control group. These women also filled in the same questionnaires, both at the start, and after one and three months. As far as the AMQ was concerned, the switchers appeared in the initial measurement to have significantly higher scores (that is to say, a worse mood) for “tiredness”, “depression”, “moodiness”, “anxiety” and “anger”, than those of the starters and the control group. Moreover, the switchers had reduced scores (that is to say, an improved mood) on the AMQ-scales during their use of Minulet compared to their initial use of oral contraceptives. The scores of the starters and the control group on the AMQ-scales remained unchanged on all three measurements. It can be concluded from this that these groups felt as well as they did before. With regard to the SIP: in the initial measurement, the switchers appeared to have less favourable scores in the SIP categories “emotional behaviour”, “social interaction”, and “alert ness” than the control group. Also, at baseline the switchers appeared to score less favourably than the starters with regard to “social interactions” only. The switchers scored more favourably on all mentioned SIP categories during their use of Minulet than at baseline. The starters and the control group continued to feel the same on all three measurements for the SIP. The results of the SIP were in accordance with the results of the AMQ. Based on this study, it was concluded that women who had not been taking OC and who started taking Minulet, continued to feel the same as they did before. Secondly, it was concluded that women who switched over from their usual OC to Minulct felt that their mood improved. INTRODUCTION Due to their active substances, in this case estrogen and progestin. oral contraceptives (OC) can have, amongst others. undesired side effects on mood. The mental side effects that have been reported over the years are irritability, depression. tiredness and anxiety. Since the introduction of OC, many studies have been performed to establish their effect on Submitted Accepted
for publication for publication
February August 3,
Copyright
0 1992 Butterworth-Heinemann
17, 1992 1992
360
Contraception
mood. A survey of studies from 1967 to 1971 was made by Cullberg [ 11. In each of the twelve investigations described, 100 to 200 women were examined by means of interviews and/or questionnaires. The largest study involved 1217 women. The mental side effects encountered in these studies were symptoms of depression and dysphotia, tiredness, neurasthenia, irritability, anxiety, nervousness and hostility [2-91. None of the experiments showed any improvement of mood resulting from the use of oral contraceptives. If the effect of estrogen is compared with that of progestin, little consistency can be found based on the above studies. According to three studies, oral contraceptives with higher levels of progestin would appear to have more mental side effects; in one study, estrogen dominance appeared to have negative side effects. Finally, two of the studies showed no difference in the effects of estrogen and progestin. Cullberg [ 1] himself found estrogen dominance and progestin dominance to have an approximately equal number of negative mental effects (symptoms of depression/dysphoria). The connection between mental symptoms and OC use was also repeatedly established after 1971 [ 10-121. In most of the studies the relationship between OC and symptoms of depression were investigated. In a number of studies, the relationship between OC and depressive mood changes was especially or most strongly found in women who already had a disposition to be depressed [ 13, 141. In order to investigate the role of genetic factors in symptoms related to OC. 7 15 monozygotic and 416 dizygotic twins, using the same OC, were interviewed [ 151. OC-related depression was influenced by genetic factors. This was also true, although to a lesser extent, for “irritability”: it was thus concluded that genetic factors can influence the mental side effects of OC USC. In connection with the findings about the relationship between depression and OC use, experiments have been performed on the biological mechanisms that could be responsible for this relationship. Important mechanisms which are influenced by OC are serotonin, noradrenalin, vitamin and endocrine metabolisms [ 161. Likewise the estrogen component can reduce the concentration of noradrenalin in the hypothalamus as well as the concentration of serotonin, resulting from pyridoxine deficiency ] 13, 14, 171. In addition, depression can be related to a rise in cerebral monoamine oxydase activity, which is caused by progestin [IS]. In 1988 a large group of women was questioned about their experiences with OC [19]. Three out of four women appeared to use contraceptives, of which the highest percentage (34%) were taking OC. With respect to complaints about the pill, 30% of the women appeared never to have been bothered by side effects. About 30% of the questioned women seemed to suffer, or to have suffered, from complaints such as headaches, weight gain and interim bleeding. Over 10% of the women were suffering, or had sul’fered in the past, from loss of libido, dejection or tiredness. Former pill users in particular complained more often about loss of libido, dejection, dizziness and high blood pressure. Since, in most cases, the complaints disappeared when these women stopped taking the pill, it can be assumed that the complaints were connected to the use of the pill. It is estimated that more than a quarter of the women taking the pill at present, are currently experiencing side effects. The consequences arc tension. ncrvousncss. lack of concentration. tiredness and dejection. In view of the fact that the side effects, which include those mental symptoms already mentioned, are related to the level of estrogen and progestin. attempts have been made to develop OC containing as little of these sexual hormones as possible. Minulet is a monophasic contraceptive, containing a new synthetic progcstin. gcstodene. The level of progestin and estrogen is low: Minulet contains per pill 75 ug gestodene and 30 ug ethinyl estradiol (EE). It is possible that this leads to fewer mental and physical complaints than would normally result from OC use. The following study was carried out to detennine the influence of Minulct on mood. For this purpose, a number of women who started taking Minulct after never having taken any oral
361
Contraception
contraceptive at all. or after having taken another type of oral contraceptive, filled in several mood questionnaires. They filled in the mood questionnaires three times: before starting to take Minulet, after one month and then again after three months. A group of women who did not take any pill at all tilled in the same questionnaires at the same times and served as a control. METHODS Subjects The women that participated varied between 16 and 45 years of age, were healthy and had a regular menstrual cycle (21-35 days) within 3 months before the study period. The exclusiotr criteria were: use of sex hormones, anticonvulsants, antibiotics/ liable to be pregnant/ severe diabetes. abnormal genital bleedings, disturbed fat metabolism, severe liver disease, breast carcinoma, herpes gestationis. otosclerosis, thrombophlebitis, estrogen-dependent neoplasm, myocardial infarction, cerebrovascular accident. The participants belonged to one of the three following groups: - 140 women who did not take any OC during the study: control group (mean age: 34 years; S.D.: 10). From these 140 subjectsone dropped out. - 200 women who had not taken an OC before and started taking Minulet: starters mean age: 22 years; S.D.: 7). From these 200 subjects,five dropped out for unknown reasons. - 370 women who were taking an OC but were switched over to Minulet: switchers (mean age: 28 years; S.D.: 6). From these 370 subjects,7 dropped out for unknown reasons. No data were collected with respect to the length that these subjects were on their usual OC prior to Minulet. Material Amsterdam Mood Questionnaire (AMQ): The AMQ [20, 211 is a questionnaire especially developed to measure mood changes (“States”). The questionnaire contains 60 items divided into 10 mood scales, each scale being assessed on the basis of scores on 6 items. The scales are: depressive, tired, elated, conscientious, shy, indifferent, moody, arrogant, angry, anxious. Two types of techniques can be distinguished. In version A, the participants are asked to judge their mood during the last few days. In version B. the mental state is determined at the moment of questioning. Version A has been used in this experiment, in order to avoid situational factors having too great an effect. The mood is indicated by circling one of the 5 possibilities of each item. These possibilities are: “absolutely not”, “a little”, “reasonably”, “good” and “very good” and they indicate the extent to which the mood described applies. The questionnaire takes about 10 minutes to complete. The nonnalisation scale of the AMQ is the Stanine scale, with u=5 and o=2. Stanine 1 corresponds to a range of the 4 percent lowest scores and stanine 9 to a range of the 4 percent highest scores observed in a normal group. Stanine 5 corresponds to the most frequently (19.8%) observed scores. Norms are based on a group of students with an average age of 22.5 years. Sickness impact Pvofile (SIP): The SIP [22] used is a Dutch adaptation of a questionnaire that was developed in the USA. The SIP aims to measure changes in behaviour and activities which. in the opinion of the respondent, are related to the state of health [23]. The SIP makes a distinction between physical and psychosocial dysfunctioning. Physical dysfunctioning covers the areas of body care and exercise, mobility and walking. The partial score for psychosocial dysfunctioning is derived from the
362
Contraception
following categories: emotional behaviour, social interaction. communication and alertness/ intellectual functioning. It is also possible to employ the categories separately. In the study concemcd, the categories “emotional behaviour”. “social interactions” and “alertness/intellectual functioning” were employed. The respondent has to circle those items applicable to her current state. The questionnaire takes about 10 minutes to complete.
The experimental group consisted of women who were visiting general practitioners, and who were either already taking an oral contraceptive or wanted to take an oral contraceptive. Both groups of women were asked to participate in the study, the import which was explained by the general practitioner concerned. The women were told that the objective of the study was to obtain information about this new OC. The participating physicians had been instructed to refrain from suggesting that Minulet would improve their mood. If a woman was interested in participating and applied lo the in- and exclusion criteria, Minulet was prescribed. Before the women started to lake Minulet, they were asked to fill in the AMQ and the SIP. Furthermore, they wcrc asked to Ii11in the AMQ and the SIP again after one month and after three months. The control group consisted 01. a random sample of women in the same age group, who wcrc asked in writing to participate in the experiment. Women who agreed to participate in the experiment filled in the AMQ and the SIP sent to them. After about one month and after about three months, they wcrc again sent Ihc AMQ and the SIP which they filled in and sent back. The scores on the AMQ and the SIP of those women not taking any OC were dctcrmincd; these women formed the control group.
The scores on the separate items of the AMQ wcrc added up, giving summated scores for each of the IO mood scales. The scales “arrogant ” . “indifferent” and “conscientious” were excluded from tbc analysis because thcsc categories had no relation to the questioning. The scores on the 3 remaining categories of the SIP were arrived at by counting. by category, the calibration units of the items circled. The scores on the AMQ and the SIP obtained in this way did not seem to be normally distributed so that nonparametric tests had to be used. For testing between groups, the Mann-Whitney U Tcsl was used, and in order to establish the differences between measurements within each separate group, the Wilcoxon Matched-Pairs Signed Ranks Test was applied (Statistical Package for the Social Scicnccs, “SPSS/PC+“). RESULTS Control group: Starters: Switchers:
no OC no OC. then Minulcr another OC, rhcn Minulct
The AMQ showed significant diffcrcnccs with regard to some mood scales. With respect to “tired”, a significant diffcrcnce was found lo exist at the initial measurement bctwcen the control group and the switchers (p = 0.0006), as well as between Ihe starters and the switchers (p = 0.004). The score on “tired” for switchers turned out to be significantly higher than for the control group and the starters. According to the AMQ. the switchers also had a significantly (p < 0.0001) reduced score on “tired” at measurement 2 in comparison with the score at the initial measurement (see Figure 1). With respect to “depressed”, the switchers scored significantly higher at the initial measurement than both the control group (p = 0.002) and the starters (p = 0.0001). The switchers’ score
363
Contraception
0
1
3
0
1
3
month
month
Figure 1. Mean tiredness scores f SE (AMQ) at the initial measurement, one
Figure 2. Mean depressive scores ? SE (AMQ) at the pre-measurement, one
month later and three months later for all
month later and three months later for all three participant groups (** p < OOOS).
three participant groups (** p < 0.005).
for depressive also appeared to be significantly reduced (p < 0.0001) at measurement 2 (using Minulet) in comparison with the score at the initial measurement (using another pill). This result is depicted in Figure 2. The scores for “moody” showed the same effect. At the initial measurement, switchers scored significantly (p = 0.002) higher than the control group and the starters (p = 0.0001). Moreover, the switchers appeared to have significantly lower scores on “moodiness” at measurement 2 (taking Minulet) than at the initial measurement (taking another pill) (see Figure 3). score score 5.55.5r tt + 7I 5.0j. \ 4.5
2.52.0 i 0
1
0
3
Figure 3.
Mean moodiness scores + SE
(AMQ) at the pre-measurement, one month later and three months later for all three participant groups (** p < 0.005).
1
3
month
month Figure
4.
Mean
anxiety
scores ?
SE
(AMQ) a1
the initial measurement, one month later and three months later for all three participant groups (* p < 0.05).
The initial measurement also showed mat the switchers scored significantly (p = 0.006) higher than the control group and the starters (p = 0.01) with respect to “anxious”. The decrease in anxiety for switchers between the initial measurement and measurement 2 also appeared to bc significant (p < 0.0001) (see Figure 4).
364
Contraception
Finally, as far as the results on the AMQ were concerned, it appeared that the switchers scored significantly (p < O.CKNl) higher in the category “angry” at the initial measurement than the control group and the starters (p = O.ooO6). The switchers scored significantly (p = O.OCO4) lower for “anger” at measurement 2 than at the initial measurement. This result can be seen in Figure 5. Similarly, significant results were not found for the other measured mood scales “shy” and “elated”. The significant results found on the SIP were in accordance with the AMQ. At the initial measurement, according to the SIP results, the switchers scored higher than the control group (p = 0.01) and the starters (p = 0.001) on “social interactions”. Switchers also scored higher than the control group in the categories “emotional behaviour” (p = 0.003) and “alertness” (p = 0.03). In the categories “emotional behaviour”. “social interactions” and “alertness”, the score of the switchers was significantly reduced at measurement 2 compared to the initial measurement (p = 0.005, p < 0.0001 and p = O.OC06. respectively). These results are depicted in Figures 6, 7 and 8. In order to get an impression of the composition of the group of switchers, the reasons for switching were traced for a subgroup consisting of 469 women. The following specific reasons were found: 152 of the 469 women (32%) mentioned interim bleeding as a reason; the invitation by their physician to participate in the Minulet study followed, having been mentioned as a reason by 90 women (19%) without complaints, but interested to take a low-dose pill. This was followed by nausea/headaches. reported by 85 women (I 8%); tender breasts, reported by 49 women (10%); weight gain, reported by 36 women (7.5%): and lastly, mental complaints, reported by 34 women (7%). These mental complaints consisted of depression, moodiness. initability and tiredness.
20.0
1.00
I
I_
.A_.__
__L
_
3
1
0
3
month
month
Figure 5. Mean anger scores f SE (AMQ)
Figure 6.
at the initial measurement. one month later and rhree months labzr for all three participant groups (** p < 0.005).
behaviour” urement.
Mean scores for _+ SE (SIP)
one month
later for all (** p < 0.005).
three
“emotional at the initial meas-
later and three months participant
groups
AMQ results per OC ,pm~p In order to make a distincnon withm the “group 01 switchers” between the types of pill being taken at the time of the initial measurement. information was gathered among a subgroup of this group of women. This furnished data on 193 women. The scores of this group, representative of the results obtained. have been split up according to the kind of pilI originally taken and
365
Contraception
90.0
(Lcor*
acore
60.0
y
70.0
L
60.0
:
30.0
-
l ; tl.
60.0
-
_
70.0
-*
controls
I
+
*t*rt*r(l
60.0
L
-
sWitCh*rs
60.0
I
40.0
i
0
I 1
20.0
3
’
0
month
month Figure 7. Mean scores for “social interaclions” k SE (SIP) at the initial measurcment. one month later and three months later for all three participant 0.05).
1
Figure 8. Mean scores for “alertness/ intellectual functioning” k SE (SIP) at the initial measurement, one month later and three months later for all three participant
groups (* p
46:359-367,
1992
USE OF A MONOPHASIC, LOW-DOSE ORAL CONTRACEPTIVE IN RELATION TO MENTAL FUNCTIONING J.B. Deijerl*, K.J. Duyn **, W.A. Jamen* and J.W. Klitsie** * Dept. of Psychophysiology, Vrije Universiteit, De Boelelaan 1111, 1081 HV Amsterdam, The Netherlands ** Wyeth Laboratoria B.V., Hoofddorp, The Netherlands
ABSTRACT The objective of the study was to evaluate the effect of Minulet, a new low-dose oral contraceptive on mood in two groups and to compare the effect with a control group of women not taking oral contraceptives (OC). The women participating were between 16 and 45 years of age. They completed the Amsterdam Mood Questionnaire (AMQ) and the Sickness Impact Profile (SIP) three times. They were filled in before treatment started, after taking Minulet for one month and then again after three months. The questionnaires were filled in by a group of 200 women who had not taken OC before (starters), and by a second group of 370 women who were already taking OC (switchers). A group of 140 women who did not use any OC during the study served as a control group. These women also filled in the same questionnaires, both at the start, and after one and three months. As far as the AMQ was concerned, the switchers appeared in the initial measurement to have significantly higher scores (that is to say, a worse mood) for “tiredness”, “depression”, “moodiness”, “anxiety” and “anger”, than those of the starters and the control group. Moreover, the switchers had reduced scores (that is to say, an improved mood) on the AMQ-scales during their use of Minulet compared to their initial use of oral contraceptives. The scores of the starters and the control group on the AMQ-scales remained unchanged on all three measurements. It can be concluded from this that these groups felt as well as they did before. With regard to the SIP: in the initial measurement, the switchers appeared to have less favourable scores in the SIP categories “emotional behaviour”, “social interaction”, and “alert ness” than the control group. Also, at baseline the switchers appeared to score less favourably than the starters with regard to “social interactions” only. The switchers scored more favourably on all mentioned SIP categories during their use of Minulet than at baseline. The starters and the control group continued to feel the same on all three measurements for the SIP. The results of the SIP were in accordance with the results of the AMQ. Based on this study, it was concluded that women who had not been taking OC and who started taking Minulet, continued to feel the same as they did before. Secondly, it was concluded that women who switched over from their usual OC to Minulct felt that their mood improved. INTRODUCTION Due to their active substances, in this case estrogen and progestin. oral contraceptives (OC) can have, amongst others. undesired side effects on mood. The mental side effects that have been reported over the years are irritability, depression. tiredness and anxiety. Since the introduction of OC, many studies have been performed to establish their effect on Submitted Accepted
for publication for publication
February August 3,
Copyright
0 1992 Butterworth-Heinemann
17, 1992 1992
360
Contraception
mood. A survey of studies from 1967 to 1971 was made by Cullberg [ 11. In each of the twelve investigations described, 100 to 200 women were examined by means of interviews and/or questionnaires. The largest study involved 1217 women. The mental side effects encountered in these studies were symptoms of depression and dysphotia, tiredness, neurasthenia, irritability, anxiety, nervousness and hostility [2-91. None of the experiments showed any improvement of mood resulting from the use of oral contraceptives. If the effect of estrogen is compared with that of progestin, little consistency can be found based on the above studies. According to three studies, oral contraceptives with higher levels of progestin would appear to have more mental side effects; in one study, estrogen dominance appeared to have negative side effects. Finally, two of the studies showed no difference in the effects of estrogen and progestin. Cullberg [ 1] himself found estrogen dominance and progestin dominance to have an approximately equal number of negative mental effects (symptoms of depression/dysphoria). The connection between mental symptoms and OC use was also repeatedly established after 1971 [ 10-121. In most of the studies the relationship between OC and symptoms of depression were investigated. In a number of studies, the relationship between OC and depressive mood changes was especially or most strongly found in women who already had a disposition to be depressed [ 13, 141. In order to investigate the role of genetic factors in symptoms related to OC. 7 15 monozygotic and 416 dizygotic twins, using the same OC, were interviewed [ 151. OC-related depression was influenced by genetic factors. This was also true, although to a lesser extent, for “irritability”: it was thus concluded that genetic factors can influence the mental side effects of OC USC. In connection with the findings about the relationship between depression and OC use, experiments have been performed on the biological mechanisms that could be responsible for this relationship. Important mechanisms which are influenced by OC are serotonin, noradrenalin, vitamin and endocrine metabolisms [ 161. Likewise the estrogen component can reduce the concentration of noradrenalin in the hypothalamus as well as the concentration of serotonin, resulting from pyridoxine deficiency ] 13, 14, 171. In addition, depression can be related to a rise in cerebral monoamine oxydase activity, which is caused by progestin [IS]. In 1988 a large group of women was questioned about their experiences with OC [19]. Three out of four women appeared to use contraceptives, of which the highest percentage (34%) were taking OC. With respect to complaints about the pill, 30% of the women appeared never to have been bothered by side effects. About 30% of the questioned women seemed to suffer, or to have suffered, from complaints such as headaches, weight gain and interim bleeding. Over 10% of the women were suffering, or had sul’fered in the past, from loss of libido, dejection or tiredness. Former pill users in particular complained more often about loss of libido, dejection, dizziness and high blood pressure. Since, in most cases, the complaints disappeared when these women stopped taking the pill, it can be assumed that the complaints were connected to the use of the pill. It is estimated that more than a quarter of the women taking the pill at present, are currently experiencing side effects. The consequences arc tension. ncrvousncss. lack of concentration. tiredness and dejection. In view of the fact that the side effects, which include those mental symptoms already mentioned, are related to the level of estrogen and progestin. attempts have been made to develop OC containing as little of these sexual hormones as possible. Minulet is a monophasic contraceptive, containing a new synthetic progcstin. gcstodene. The level of progestin and estrogen is low: Minulet contains per pill 75 ug gestodene and 30 ug ethinyl estradiol (EE). It is possible that this leads to fewer mental and physical complaints than would normally result from OC use. The following study was carried out to detennine the influence of Minulct on mood. For this purpose, a number of women who started taking Minulct after never having taken any oral
361
Contraception
contraceptive at all. or after having taken another type of oral contraceptive, filled in several mood questionnaires. They filled in the mood questionnaires three times: before starting to take Minulet, after one month and then again after three months. A group of women who did not take any pill at all tilled in the same questionnaires at the same times and served as a control. METHODS Subjects The women that participated varied between 16 and 45 years of age, were healthy and had a regular menstrual cycle (21-35 days) within 3 months before the study period. The exclusiotr criteria were: use of sex hormones, anticonvulsants, antibiotics/ liable to be pregnant/ severe diabetes. abnormal genital bleedings, disturbed fat metabolism, severe liver disease, breast carcinoma, herpes gestationis. otosclerosis, thrombophlebitis, estrogen-dependent neoplasm, myocardial infarction, cerebrovascular accident. The participants belonged to one of the three following groups: - 140 women who did not take any OC during the study: control group (mean age: 34 years; S.D.: 10). From these 140 subjectsone dropped out. - 200 women who had not taken an OC before and started taking Minulet: starters mean age: 22 years; S.D.: 7). From these 200 subjects,five dropped out for unknown reasons. - 370 women who were taking an OC but were switched over to Minulet: switchers (mean age: 28 years; S.D.: 6). From these 370 subjects,7 dropped out for unknown reasons. No data were collected with respect to the length that these subjects were on their usual OC prior to Minulet. Material Amsterdam Mood Questionnaire (AMQ): The AMQ [20, 211 is a questionnaire especially developed to measure mood changes (“States”). The questionnaire contains 60 items divided into 10 mood scales, each scale being assessed on the basis of scores on 6 items. The scales are: depressive, tired, elated, conscientious, shy, indifferent, moody, arrogant, angry, anxious. Two types of techniques can be distinguished. In version A, the participants are asked to judge their mood during the last few days. In version B. the mental state is determined at the moment of questioning. Version A has been used in this experiment, in order to avoid situational factors having too great an effect. The mood is indicated by circling one of the 5 possibilities of each item. These possibilities are: “absolutely not”, “a little”, “reasonably”, “good” and “very good” and they indicate the extent to which the mood described applies. The questionnaire takes about 10 minutes to complete. The nonnalisation scale of the AMQ is the Stanine scale, with u=5 and o=2. Stanine 1 corresponds to a range of the 4 percent lowest scores and stanine 9 to a range of the 4 percent highest scores observed in a normal group. Stanine 5 corresponds to the most frequently (19.8%) observed scores. Norms are based on a group of students with an average age of 22.5 years. Sickness impact Pvofile (SIP): The SIP [22] used is a Dutch adaptation of a questionnaire that was developed in the USA. The SIP aims to measure changes in behaviour and activities which. in the opinion of the respondent, are related to the state of health [23]. The SIP makes a distinction between physical and psychosocial dysfunctioning. Physical dysfunctioning covers the areas of body care and exercise, mobility and walking. The partial score for psychosocial dysfunctioning is derived from the
362
Contraception
following categories: emotional behaviour, social interaction. communication and alertness/ intellectual functioning. It is also possible to employ the categories separately. In the study concemcd, the categories “emotional behaviour”. “social interactions” and “alertness/intellectual functioning” were employed. The respondent has to circle those items applicable to her current state. The questionnaire takes about 10 minutes to complete.
The experimental group consisted of women who were visiting general practitioners, and who were either already taking an oral contraceptive or wanted to take an oral contraceptive. Both groups of women were asked to participate in the study, the import which was explained by the general practitioner concerned. The women were told that the objective of the study was to obtain information about this new OC. The participating physicians had been instructed to refrain from suggesting that Minulet would improve their mood. If a woman was interested in participating and applied lo the in- and exclusion criteria, Minulet was prescribed. Before the women started to lake Minulet, they were asked to fill in the AMQ and the SIP. Furthermore, they wcrc asked to Ii11in the AMQ and the SIP again after one month and after three months. The control group consisted 01. a random sample of women in the same age group, who wcrc asked in writing to participate in the experiment. Women who agreed to participate in the experiment filled in the AMQ and the SIP sent to them. After about one month and after about three months, they wcrc again sent Ihc AMQ and the SIP which they filled in and sent back. The scores on the AMQ and the SIP of those women not taking any OC were dctcrmincd; these women formed the control group.
The scores on the separate items of the AMQ wcrc added up, giving summated scores for each of the IO mood scales. The scales “arrogant ” . “indifferent” and “conscientious” were excluded from tbc analysis because thcsc categories had no relation to the questioning. The scores on the 3 remaining categories of the SIP were arrived at by counting. by category, the calibration units of the items circled. The scores on the AMQ and the SIP obtained in this way did not seem to be normally distributed so that nonparametric tests had to be used. For testing between groups, the Mann-Whitney U Tcsl was used, and in order to establish the differences between measurements within each separate group, the Wilcoxon Matched-Pairs Signed Ranks Test was applied (Statistical Package for the Social Scicnccs, “SPSS/PC+“). RESULTS Control group: Starters: Switchers:
no OC no OC. then Minulcr another OC, rhcn Minulct
The AMQ showed significant diffcrcnccs with regard to some mood scales. With respect to “tired”, a significant diffcrcnce was found lo exist at the initial measurement bctwcen the control group and the switchers (p = 0.0006), as well as between Ihe starters and the switchers (p = 0.004). The score on “tired” for switchers turned out to be significantly higher than for the control group and the starters. According to the AMQ. the switchers also had a significantly (p < 0.0001) reduced score on “tired” at measurement 2 in comparison with the score at the initial measurement (see Figure 1). With respect to “depressed”, the switchers scored significantly higher at the initial measurement than both the control group (p = 0.002) and the starters (p = 0.0001). The switchers’ score
363
Contraception
0
1
3
0
1
3
month
month
Figure 1. Mean tiredness scores f SE (AMQ) at the initial measurement, one
Figure 2. Mean depressive scores ? SE (AMQ) at the pre-measurement, one
month later and three months later for all
month later and three months later for all three participant groups (** p < OOOS).
three participant groups (** p < 0.005).
for depressive also appeared to be significantly reduced (p < 0.0001) at measurement 2 (using Minulet) in comparison with the score at the initial measurement (using another pill). This result is depicted in Figure 2. The scores for “moody” showed the same effect. At the initial measurement, switchers scored significantly (p = 0.002) higher than the control group and the starters (p = 0.0001). Moreover, the switchers appeared to have significantly lower scores on “moodiness” at measurement 2 (taking Minulet) than at the initial measurement (taking another pill) (see Figure 3). score score 5.55.5r tt + 7I 5.0j. \ 4.5
2.52.0 i 0
1
0
3
Figure 3.
Mean moodiness scores + SE
(AMQ) at the pre-measurement, one month later and three months later for all three participant groups (** p < 0.005).
1
3
month
month Figure
4.
Mean
anxiety
scores ?
SE
(AMQ) a1
the initial measurement, one month later and three months later for all three participant groups (* p < 0.05).
The initial measurement also showed mat the switchers scored significantly (p = 0.006) higher than the control group and the starters (p = 0.01) with respect to “anxious”. The decrease in anxiety for switchers between the initial measurement and measurement 2 also appeared to bc significant (p < 0.0001) (see Figure 4).
364
Contraception
Finally, as far as the results on the AMQ were concerned, it appeared that the switchers scored significantly (p < O.CKNl) higher in the category “angry” at the initial measurement than the control group and the starters (p = O.ooO6). The switchers scored significantly (p = O.OCO4) lower for “anger” at measurement 2 than at the initial measurement. This result can be seen in Figure 5. Similarly, significant results were not found for the other measured mood scales “shy” and “elated”. The significant results found on the SIP were in accordance with the AMQ. At the initial measurement, according to the SIP results, the switchers scored higher than the control group (p = 0.01) and the starters (p = 0.001) on “social interactions”. Switchers also scored higher than the control group in the categories “emotional behaviour” (p = 0.003) and “alertness” (p = 0.03). In the categories “emotional behaviour”. “social interactions” and “alertness”, the score of the switchers was significantly reduced at measurement 2 compared to the initial measurement (p = 0.005, p < 0.0001 and p = O.OC06. respectively). These results are depicted in Figures 6, 7 and 8. In order to get an impression of the composition of the group of switchers, the reasons for switching were traced for a subgroup consisting of 469 women. The following specific reasons were found: 152 of the 469 women (32%) mentioned interim bleeding as a reason; the invitation by their physician to participate in the Minulet study followed, having been mentioned as a reason by 90 women (19%) without complaints, but interested to take a low-dose pill. This was followed by nausea/headaches. reported by 85 women (I 8%); tender breasts, reported by 49 women (10%); weight gain, reported by 36 women (7.5%): and lastly, mental complaints, reported by 34 women (7%). These mental complaints consisted of depression, moodiness. initability and tiredness.
20.0
1.00
I
I_
.A_.__
__L
_
3
1
0
3
month
month
Figure 5. Mean anger scores f SE (AMQ)
Figure 6.
at the initial measurement. one month later and rhree months labzr for all three participant groups (** p < 0.005).
behaviour” urement.
Mean scores for _+ SE (SIP)
one month
later for all (** p < 0.005).
three
“emotional at the initial meas-
later and three months participant
groups
AMQ results per OC ,pm~p In order to make a distincnon withm the “group 01 switchers” between the types of pill being taken at the time of the initial measurement. information was gathered among a subgroup of this group of women. This furnished data on 193 women. The scores of this group, representative of the results obtained. have been split up according to the kind of pilI originally taken and
365
Contraception
90.0
(Lcor*
acore
60.0
y
70.0
L
60.0
:
30.0
-
l ; tl.
60.0
-
_
70.0
-*
controls
I
+
*t*rt*r(l
60.0
L
-
sWitCh*rs
60.0
I
40.0
i
0
I 1
20.0
3
’
0
month
month Figure 7. Mean scores for “social interaclions” k SE (SIP) at the initial measurcment. one month later and three months later for all three participant 0.05).
1
Figure 8. Mean scores for “alertness/ intellectual functioning” k SE (SIP) at the initial measurement, one month later and three months later for all three participant
groups (* p
