Complementary Therapies in Medicine (2014) 22, 203—211
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevierhealth.com/journals/ctim
A randomized controlled trial of the effects of a stress management programme during pregnancy Christina Tragea a, George P. Chrousos a,b, Evangelos C. Alexopoulos a,1, Christina Darviri a,∗,1 a
Postgraduate Course Stress Management and Health Promotion, School of Medicine, University of Athens, Biomedical Research Foundation, Academy of Athens, Soranou Ephessiou Str. 4, GR-115-27 Athens, Greece b First Department of Pediatrics, Children’s Hospital Aghia Sofia, School of Medicine, University of Athens, Thivon & Papadiamantopoulou Str., GR-115-27 Athens, Greece Available online 30 January 2014
KEYWORDS Anxiety; Stress; Pregnancy; Relaxation; Stress management
∗ 1
Summary Background: Prenatal maternal stress is associated with adverse birth outcomes. Relaxation techniques might be effective in reducing stress during that period. The purpose of this study was to evaluate the effects of applied relaxation in reducing anxiety and stress in pregnant women in their second trimester, as well as raising their sense of control. Also we expected to see a difference in some lifestyle factors associated with stress. A randomized control trial with a prospective pretest—posttest experimental design was used. Methods: Sixty primigravida women in their second trimester were assigned randomly to receive a 6-week stress management programme (N = 31) (relaxation breathing and progressive muscle relaxation, RB-PMR, twice a day) or not (N = 29). Self-reported validated measures were used to evaluate perceived stress, health locus of control and anxiety at baseline and at the end of the 6-weeks follow-up. Results: The results of the study demonstrated significant benefits from the use of the techniques in the psychological state of the pregnant women. The systematic implementation of the proposed relaxation techniques contributed in the reduction of perceived stress (mean change −3.23, 95% CI: −4.29 to −0.29) and increased the sense of control (mean change 1.99, 95% CI: 0.02—3.7). Conclusion: The findings suggest beneficial effects of relaxation on reducing perceived stress as well as increment of sense of control in pregnant women. The results of this study support the claim that training in the proposed relaxation techniques may constitute an ideal, nonpharmaceutical, intervention that can promote well-being, at least during pregnancy. Longer studies will be necessary in the future, in order to examine the long-term effects of relaxation techniques. © 2014 Elsevier Ltd. All rights reserved.
Corresponding author at: Soranou Ephessiou Str. 4, GR-115-27 Athens, Greece. Tel.: +30 210 6597644; fax: +30 210 6597644. E-mail address: [email protected] (C. Darviri). These authors contributed equally and share last authorship.
0965-2299/$ — see front matter © 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ctim.2014.01.006
204
Introduction Pregnancy for many women it may be particularly stressful, significantly affecting their quality of life, especially for the primigravida. Many factors have been associated with prenatal psychological stress. They generally include the sum of all arising demands, changes in body fitness, the shifting roles and the differentiation in interpersonal relationships with significant others.1 Although, these changes may be stressful, pregnancy is something most women choose to experience and as a result, birth is a welcome event usually accompanied by great enthusiasm. Moreover, in contrast to other stressful life events, pregnancy is a finite event with a specific duration and outcome which, in most cases, can be predicted with great accuracy.2 A significant number of studies in animals and humans have demonstrated that high maternal stress and mood disorders during pregnancy are associated with a variety of negative repercussions for both mother and foetus, including premature birth,3—7 higher risk of developing hypertension and preeclampsia,8 problems in lactation9 and increased vulnerability to mood disorders and autoimmune phenomena during the postpartum period.9 Furthermore, maternal stress has been shown to adversely affect the foetus’ development in the womb (brain development and neurological damage)10—15 and it is associated with the low birth weight of the infant, the small head circumference, the low Apgar score16—18 and with increased risk of miscarriage or stillbirth.19 It has also been demonstrated that maternal stress may later affect the relationship between mother and child as well as the child’s adult life (autism, schizophrenia, ADHD)20—25 and predisposes a difficult childbirth and the emergence of depressive symptoms during the postpartum period.26—29 In physiology, increased levels of stress, anxiety and depressed mood have been shown to correlate with changes in various parameters mainly through the activation of hypothalamic—pituitary—adrenal axis (HPA axis) and the sympathetic nervous system.30 Various hormones, including corticotropin hormone (CRH), adrenocorticotropic hormone (ACTH), cortisol and norepinephrine (NE), are released in large quantities in the bloodstream.30 How all these changes in physiological parameters may lead to adverse effects is not yet well understood. Furthermore, although research has focused more on the correlation between stress and possible complications in pregnancy, there is no significant bibliography demonstrating the usefulness of specific relaxation techniques in reducing stress during this period.31 It is anticipated that the systematic implementation of relaxation techniques may lead to important positive physiological changes which include a decrease in metabolism, oxygen consumption, cardiac function, blood pressure, breathing rate, and brain activity as well as to a significant increase in well-being.32 In the last decade, complementary-alternative treatments have been largely incorporated in routine medical practices, including pregnancy.33 Studies on the effect of relaxation techniques during pregnancy reveal a variety of significant effects.31 The systematic implementation of relaxation techniques has shown to reduce the likelihood of premature or delayed birth, to increase the infant’s body weight and to reduce the likelihood of a Caesarean
C. Tragea et al. section and birth interventions.34—38 Relaxation techniques that have been studied in pregnancy include most frequently yoga, hypnotherapy (guided hypnotherapy imagery script) and meditation with promising results in the reduction of stress, perceived stress, the reduction of various physical symptoms and the promotion of good mood and well-being.37—40 Applied relaxation, which is a combination of diaphragmatic breathing and muscle relaxation, has also been found to reduce anxiety and perceived stress and have other positive effects.34—36,41,42 There are limited data to support the efficacy of applied relaxation for stress reduction during pregnancy. This is the first study in Greek women, to examine change in perceived stress, anxiety and sense of control following a six week programme of a combination of two relaxation techniques (diaphragmatic breathing and progressive muscle relaxation). The purpose of this study was to evaluate the effects of applied relaxation (combination of diaphragmatic breathing and muscle relaxation) in reducing anxiety and stress in pregnant women during their second trimester, as well as raising their sense of control. By implementing this low cost non-pharmaceutical randomized controlled intervention, we expected a significant reduction in state-trait anxiety levels and in perceived stress and an increase in the sense of health control by participants in the intervention group compared with the control group. Also we expected to see a difference in some lifestyle factors associated with stress.
Method The design of the study was a randomized control trial (RCT), which included an experimental trial before and after the intervention (prospective pre-test—post-test experimental design). The main study hypotheses were as follows: (1) state anxiety scores of the experimental group will be significantly reduced compared with the control groups scores, after applied relaxation training, (2) trait anxiety scores of the experimental group will be significantly reduced compared with the control group’s scores, after applied relaxation training, (3) perceived stress scores of the experimental group will be significantly reduced compared with the control group’s scores, after applied relaxation training, and (4) sense of control will be significantly increased compared with the control group’s scores, after applied relaxation training. The sample group for this study was a convenience sample. It consisted of pregnant women invited from private obstetricians practice and the Attikon maternity hospital in Attica, more specifically from the department of Foetal Medicine. The study was accredited from the medical school of Athens and the Scientific Committee of Attikon hospital (registration number: UoAMedPR-4716-180211-23). The admission criteria for the study required that participants were primigravida, in the second trimester of their pregnancy (14—28 weeks gestation), aged 20—40 years, with singleton pregnancy and no prior history of medical complications. Also, they were required to understand Greek. Exclusion criteria were the therapeutic use of drugs for
Effects of a stress management programme medical complications and the systematic use of other relaxation techniques. The criterion for selection and monitoring of pregnant women in their second trimester was chosen because during the second trimester, pregnant women have organized the programme of maternity care they will follow. Additionally, there is enough time to implement a stress management programme (such as the specific which lasts 6 weeks), as well as enough time to evaluate the results before the major event of birth in another context (longer follow up) after birth. The study took place in Attica, from October 2010 to June 2011.The women, who participated in the study, after being informed about the purpose and process of the study, provided written consent prior to their participation in the study. The women who agreed to participate in the study were divided into two groups by randomization. The method of randomization used, resulted from an online random number generator based on cosmic rays (random.org). The two groups were the intervention group, which would implement the proposed relaxation techniques and the control group which would be given general guidelines in order to improve their lives and reduce stress. Both groups initially received a brief lecture on stress and the effectiveness of management techniques, while they were also given a brochure on stress which included the meaning of stress, its pathophysiology, the factors causally associated with it and the major consequences associated with pregnancy. They were also given brochures on the benefits of diet and exercise tailored for that specific period. In our study two types of interventions, diaphragmatic breathings and progressive muscle relaxation were used and were accounted for the observed effects. The education and counselling regarding healthy diet and regular exercise that both groups received in the beginning of the study consists a common practice in Greek maternity hospitals that is given to all pregnant women. The nutrition brochure included the fundamentals of the Mediterranean diet. These include regular small meals ensuring variety within the diet, with particular attention to breakfast. They were also given instructions on the frequency, quantity and type of food for each meal, with the aim of maintaining a stable body weight and receiving all essential nutrients tailored to the specific period of pregnancy they are going through. The brochure on exercise highlighted the importance of exercise and the fact that it can act effectively just like any other therapeutic agent. It included the main benefits of exercise, especially during pregnancy. A course of moderate exercise was recommended as well as activities generally recognized as safe during this period (walking, swimming, running and aerobics). The intervention group was given an audio cd (lasting 20 min with clear instructions) with the proposed relaxation techniques (diaphragmatic breathing exercises in combination with muscle relaxation exercises). This group was asked to apply the techniques as often as possible (1—2 times a day for 6 weeks). Relaxation through diaphragmatic breathing is accomplished by taking deep diaphragmatic breaths followed by
205 prolonged slow exhalation. The second part of the audio cd included progressive muscle relaxation (PMR) exercises. The process of PMR is designed to teach the way to relax body muscles through a two-step process. Initially, active contraction is applied to specific muscle groups. Then, the intensity stops and each individual turn her attention to observe how muscles relax after prolonged contraction. Moreover, they were given a diary to record and control the frequency of relaxation techniques, which they were asked to fill out each time they applied the relaxation techniques. Additionally, the intervention group was given a brochure containing instructions about the importance of adopting a healthy lifestyle through a routine and proper time management. Particular emphasis was placed on the importance of maintaining a programme in everyday life and time management, as well as adopting a routine which seems to have much effect in reducing stress and promoting good health. Monitoring the pregnant women was performed weekly by telephone communication or a personal meeting, when deemed necessary, throughout the duration of the programme, where the frequency of relaxation techniques was tested and the benefits of the techniques and any queries were discussed. For the sake of similarity between the two groups, there was a telephone communication once every week with the control group too. At the end of the six week programme, women in the control group were awarded by introducing the techniques accompanied with the relaxation cd. Data for the study were collected through self-report questionnaires that included the measurement of stress, anxiety and sense of control. Furthermore, a questionnaire was given, developed by the research team, measuring health and lifestyle — socio-demographic data and a few questions about the exact gestational age and the main concerns of pregnant women during that period. The questionnaires were handed out to the participants on the first meeting, as well as at the beginning of the programme and at the end after 6 weeks in both groups (intervention and control). Those participants who did not return completed questionnaires at the first meeting were called to return them after a period of three (3) days. The questionnaires used in this study were the following. The Perceived Stress Scale (PSS) is the most commonly used psychological tool for measuring the perception of stress. It includes a 5-point Likert-type scale ranging from never to very often. It is a tool for measuring the degree to which the situations in a person’s life are assessed as stressful. The questions are designed to demonstrate how unpredictable and uncontrollable the respondent considers the situations she experiences. The scale includes a series of 14 questions measuring perceived stress. The content of the questions is relative to the feelings and thoughts of the individual during the last month.43 A translated and validated version of PSS in Greek was utilized.44 The questionnaire of measurement of everyday life, lifestyle and health includes questions relating to the characteristics of sleep, eating habits, medical history, medical history, general health status, exercise and a list of various stress symptoms. This list does not constitute a psychometric tool. It consists of various symptoms associated in some degree with stress.
206
C. Tragea et al. 130 people were assessed
58 Not eligible 8 Declined to participate
85 Eligible
50 excluded 85 Randomized
44 Randomized to the Intervention
41 Randomized to the control group
6 weeks follow up
6 weeks follow up
4 Unable to contact
5 Unable to contact
3 Discontinued (for personal reasons)
4 Discontinued for personal reasons
1 stillbirth
3 Had medical complications
1 death of husband
12 Total
3 Had medical complications 1 Gave birth premature 13 Total 29 Included in primary analyses 31 Included in primary
60 participants
Figure 1
Flowchart of participants.
The Health Locus of Control, on a theoretical level, evaluates how people treat their health, as a result of their own behaviour (internal control centre for health), or as a result of other factors such as luck (external control centre for health). This is a self-report questionnaire, of 18 sentences. Respondents choose, on a scale of six, the degree to which each sentence represents them. Three subscales are calculated. The first measures the degree to which an individual feels they have control over their health (HLC1). The second measures the degree to which an individual feels that others are in control of their health (HLC2). The third subscale measures the degree to which an individual feels their health is a matter of luck (HLC3).45 A translated and criteria-cued version of HLC in Greek was utilized. The State-Trait Anxiety. The purpose of this questionnaire is to measure anxiety. In this scale, anxiety as a personality trait is differentiated from anxiety caused by
external factors. It consists of 40 topics, of which the first 20 refer to the respondent’s emotional state at the time of completing the questionnaire (i.e. transient stress as a result of the present situation) while the remaining 20 refer to the individual’s emotional state in general (i.e. constant stress as a personality trait). The evaluation was conducted according to the representativeness of the content of proposals for each respondent, on the basis of a 4-point Likert-type scale ranging from never to very much.46,47 The researchers used a version of STAI, translated and adapted into Greek. For the description of the qualitative features of the present study, absolute (number of observations) and relative (percentage) frequencies were used. The description of quantitative characteristics was based on the calculation of the median and the interquartile range — IQR. The statistical evaluation of the relationship between the group
Demographic and baseline characteristics of the participants at pre-intervention.
Age (years; median (IQR)) Gestational age (weeks; median (IQR)) Married (n, %) Planned pregnancy (n, %) Live in Greece (n, %) Education (years; median (IQR)) Education (n, %)
Intervention group (N = 31)
Control group (N = 29)
32.0 (28.0, 35.0) 17.0 (14.0, 21.0)
32.0 (29.0, 35.0) 17.0 (14.0, 20.0) 26 20 28 16.0
25 24 30 16.0
(80.6) (77.4) (96.8) (14.16)
(89.7) (69.0) (96.6) (14, 16)
Elementary High school Greater than diploma Income (in euros; n, %)
0 (0) 8 (25.8) 23 (74.2)
1 (3.4) 7 (24.1) 21 (72.3)
Low 1500
Intervention group (N = 31)
Control group (N = 29)
PSS-14 score (median (IQR)) A-state score (median (IQR))
23.0 (18.0, 27.0) 38.0 (35.0, 42.0)
22.0 (16.0, 27.0) 40.0 (30.0, 52.0)
A-trait score (median (IQR)) HLC1 score (median (IQR)) HLC2 score (median (IQR)) HLC3 score (IQR)
37.0 26.0 22.0 12.0
36.0 26.0 19.0 15.0
Satisfied with sleep quality (n, %) Smoking (n, %) Repressed anger (n, %) Sleep disorders (n, %) Complete control of things (n, %) Frequently Very frequently Lack of attention/inability to concentrate (n, %) Reduced memory capacity (n, %) Reported statement of stress (n, %) Tension headaches (n, %)
(31.0, (23.0, (18.0, (11.0,
42.0) 29.0) 26.0) 19.0)
(30.0, (24.0, (16.0, (11.0,
21 (67.7)
21 (72.4)
2 5 12 11
4 11 13 10
(6.7) (16.1) (38.7) (35.5)1 (3.2)
45.0) 29.0) 22.0) 20.0)
Effects of a stress management programme
Table 1
(13.8) (37.9) (44.8) (34.5)3 (10.3)
7 (22.6)
7 (24.1)
8 (25.8)
9 (31.0)
22 (71.0)
22 (75.9)
11 (35.5)
9 (31.0)
Interquartile range — IQR. PSS, Perceived Stress Scale; MHLC, Multidisciplinary Health Locus of Control (1, internal; 2, external; 3, chance); STAI, Anxiety; State-Trait Anxiety Inventory (S, state; T, trait).
207
208 (intervention—control) and the qualitative characteristics was based on Pearson’s Chi-squared (2 ) test. The statistical evaluation of the quantitative characteristics (by means of change: after minus before) according to group was conducted by applying the analysis of covariance (ANCOVA), after adjusting for baseline scores. Correlation between interval measures was calculated with Spearman’s rho coefficient. Level of significance was p 0.05. For the analysis we used the SPSS 17.0 software.
Results A flow-chart of the study appears in Fig. 1. A total of 142 women were initially evaluated during the aforementioned time period (October 2010 to June 2011). Of these, 52 individuals did not meet the entry criteria for the study, 10 were in the first trimester of their pregnancy, 13 were in the third trimester of their pregnancy, 18 were multiparous, 2 did not understand Greek, 5 systematically received medication for medical complications, 2 were carrying twins and 2 systematically applied other relaxation techniques. Of the remaining 90 participants, 5 refused to participate in the study for personal reasons. The number of pregnant women who agreed to participate in the study amounted to 85 individuals. Of the 44 individuals allocated in the intervention group, 3 were excluded because they experienced complications in their pregnancy and required systematic medication, 1 woman gave birth prematurely, 1 woman terminated pregnancy for medical reasons, 1 woman experienced an acutely stressful event (death of partner), 3 refused to continue the management programme and were excluded, while for another 4 phone communication was impossible. Of the 41 individuals allocated in the control group, 3 women had medical complications, 4 discontinued their participation for personal reasons while for the other 5 phone communication was impossible. As noted, both groups had similar dropouts. Subtle discrepancy was noted for 3 women with medical complications in the control group and 6 from the intervention group (3 with medical complications and 3 excluded due to stress, premature birth and stillbirth). The number of pregnant women who completed the stress management programme was, ultimately, 60 people. The median age (MIP) for women was 32 (29.35) years. The median pregnancy week on entry was week 17 (14.21). 85% of the participants were married and 73.3% stated that their pregnancy was planned. Also, 96.7% stated they had spent most their life in Greece. 73.3% had completed higher education while the majority reported a moderate socioeconomic level (n = 35) (Table 1). Baseline characteristics of the participants at preintervention are also showed in Table 1 and no significant differences were noted. At the end of the study, it was observed that the percentage of pregnant women who reported being happy with their sleep was higher amongst individuals in the intervention group (87.1%) in comparison with the control group (62.1%), (p-value = 0.025). Also, significantly (p < 0.05) more women in the control group reported continuation of smoking (13.8%), a sense of repression and
C. Tragea et al. anger (20.7%), sleep disorders (41.4%) in comparison with those in the intervention group (0%, 0% and 16.1%, respectively). Finally, it was observed that pregnant women in the intervention group (74.2%) reported a higher percentage of a frequent sense of complete control of things during the final month, in comparison with those in the control group (41.4%), at the end of the study (p-value = 0.018) (Table 2). At the end of the study, significant differences were observed in the scores of the questionnaires examined (Table 3). PSS-14 scores were significantly reduced (mean change −3.23, 95% CI: −4.29 to −0.29) compared to controls. Also, significant improvement for internal health locus of control (mean change 1.99, 95% CI: 0.02—3.7) was noted for women in the intervention group. No other significant difference was noted for the other variables (Table 3). Most women in the intervention group practiced the relaxation techniques for at least once per day (42 times) out of the 84 proposed during the 6 week period, while ten of them practiced the relaxation CD 84 times in total during the six week programme. Correlations between compliance with the stress management technique and mean outcome changes revealed that compliance was significant only for internal health locus of control change (Spearman’s rho = 0.411, p value 0.02).
Discussion The present study examined the effectiveness of two relaxation techniques (progressive muscle relaxation and diaphragmatic breathing) during pregnancy. The results of the study demonstrate significant benefits from the use of the techniques in the psychological state of the pregnant women. The systematic implementation of the proposed relaxation techniques contributed in the reduction of perceived stress and anxiety and increased the sense of internal control. Also we found changes in many lifestyle factors associated with stress. Our results replicate previous studies both for lifestyle factors (such as sleep, quality of life, etc.) and psychosocial ones (such as stress, health locus of control, anger, etc.). In specific, few studies confirm our positive results on quality of sleep, less sleep disorders, less feeling of repression and anger which may reflect the benefit of lowering stress and anxiety levels.36,48 Also, one study, using a big sample of 110 pregnant women, significantly reduced anxiety and perceived stress.34 PMR along with other techniques can also reduce stress levels during pregnancy, baseline heart rate and blood pressure, which is related with stress levels.36,41 More interestingly, stress management can beneficially affect pregnancy-related outcomes such as low birth weight, frequency of Caesarean sections and length of pregnancy.35,42 Direct comparisons of our study with the aforementioned are not possible due to different tool utilization for measuring psychosocial variables. We are aware that the results found in our study may have been partially differentiated in the intervention group due to the existence of a trainer—consultant. It is also possible that the more frequent phone and/or personal contact with the trainer in the experimental group may have helped to reduce stress and anxiety, apart from the implementation
Effects of a stress management programme Table 2
Comparisons of lifestyle characteristics of the participants at post intervention.
Satisfied with sleep quality (n, %) Smoking (n, %) Repressed anger (n, %) Sleep disorders (n, %) Complete control of things (n, %) Lack of attention/inability to concentrate (n, %) Impaired memory capacity (n, %) Reported statement of stress (n, %) Tension headaches (n, %) *
209
Intervention group (N = 31)
Control group (N = 29)
27 (87.1) 0 0 5 (16.1) 23 (74.2%) 8 (25.8) 9 (29.0) 17 (54.8) 4 (12.9)
18 (62.1) 4 (13.8) 6 (20.7) 12 (41.4) 12 (41.4%) 3 (10.3) 5 (17.2) 17 (58.6) 6 (20.7)
0.025* 0.046* 0.028* 0.023* 0.018* 0.093 0.097 0.768 0.14
Level of significance p < 0.05.
of the techniques. In order to minimize we had organized weekly telephone communication with women of both groups during which concerns and queries were discussed. Thus sense of offer was fostered in all women. Moreover it was not possible to monitor and control adherence to the stress management programme by the participants, besides the diary given for recording the frequency of technique implementation. Most of the participants of this study were married, with a desired pregnancy and had had several years of education; therefore results may not be generalized and may differ in unmarried women, with an unintended pregnancy and lower level of education. The profile of the participants in our study (desired pregnancy, well educated) might have influenced their adherence to and expectation by the proposed relaxation techniques at home in order to ensure a healthy pregnancy, as another study has shown.49 Another limitation is the per protocol analysis, although similar dropouts were noted in both groups. Moreover one more limitation of the study involves the measurement of pregnancy specific stress by using questionnaires designed to measure general stress. Many research groups distinguish pregnancy specific stress,50,51 proposing specific tool for measuring stress (such as the Prenatal Distress Questionnaire, the Pregnancy Related Anxiety Questionnaire and the Pregnancy Experience Scale) as more powerful and reliable
Table 3
compared with others not related to pregnancy, especially in predicting premature birth.2,51—53 The results suggest that stress management during pregnancy resulted in a significantly decline in anxiety and perceived stress. Also a positive affect in the sense of control was also found. Despite the limitations, our results support that training in the proposed simple, cheap and easily implemented relaxation techniques may constitute an ideal, non-pharmaceutical, intervention that can promote well-being, at least during mid-pregnancy. This pilot study suggested that stress management during pregnancy is feasible and preliminary evidence from this sample supports its potential efficacy in this areas, particularly if started early in the pregnancy. Lengthier follow-up studies will be necessary in the future, in order to examine the long-term effects of relaxation techniques by using pregnancy specific stress tools. Based on these preliminary results and given the poor literature in the subject, we strongly support that health professionals — and more specifically midwifes and obstetricians — should be educated in relaxation techniques, and in turn to encourage women to learn and implement relaxation, in order to study further the possible benefits in pregnant mental health, in pregnancy outcomes and childrearing.
Mean changes (95% confidence interval) in scores (end—beginning of study) according to each group.
PSS-14 score A-state score A-trait score HLC1 score HLC2 score HLC3 score
Intervention group
Control group
Mean (95% CI)
Mean (95% CI)
−3.68 −3.53 −3.84 1.71 −0.32 −0.23
−0.45 −2.03 −1.55 −0.28 1.48 −0.55
[±1.80] [±2.80] [±1.43] [±0.98] [±1.43] [±1.25]
[±1.80] [±2.88] [±2.46] [±1.64] [±1.07] [±1.45]
Between group difference 95% CI
−3.23 [−4.29 to −0.29]* −1.5 [−2.7 to 1.7] −2.29 [−4.9 to 0.3] 1.99 [0.02 to 3.7]* 1.8 [−0.7 to 3.5] −0.32 [−2.12 to 2.01]
ANCOVA using baseline scores as covariate. PSS, Perceived Stress Scale; MHLC, Multidisciplinary Health Locus of Control (1, internal; 2, external; 3, chance); STAI, Anxiety; State-Trait Anxiety Inventory (S = state, T = trait). * 95% confidence interval (CI) that do not contain the value zero (p value
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevierhealth.com/journals/ctim
A randomized controlled trial of the effects of a stress management programme during pregnancy Christina Tragea a, George P. Chrousos a,b, Evangelos C. Alexopoulos a,1, Christina Darviri a,∗,1 a
Postgraduate Course Stress Management and Health Promotion, School of Medicine, University of Athens, Biomedical Research Foundation, Academy of Athens, Soranou Ephessiou Str. 4, GR-115-27 Athens, Greece b First Department of Pediatrics, Children’s Hospital Aghia Sofia, School of Medicine, University of Athens, Thivon & Papadiamantopoulou Str., GR-115-27 Athens, Greece Available online 30 January 2014
KEYWORDS Anxiety; Stress; Pregnancy; Relaxation; Stress management
∗ 1
Summary Background: Prenatal maternal stress is associated with adverse birth outcomes. Relaxation techniques might be effective in reducing stress during that period. The purpose of this study was to evaluate the effects of applied relaxation in reducing anxiety and stress in pregnant women in their second trimester, as well as raising their sense of control. Also we expected to see a difference in some lifestyle factors associated with stress. A randomized control trial with a prospective pretest—posttest experimental design was used. Methods: Sixty primigravida women in their second trimester were assigned randomly to receive a 6-week stress management programme (N = 31) (relaxation breathing and progressive muscle relaxation, RB-PMR, twice a day) or not (N = 29). Self-reported validated measures were used to evaluate perceived stress, health locus of control and anxiety at baseline and at the end of the 6-weeks follow-up. Results: The results of the study demonstrated significant benefits from the use of the techniques in the psychological state of the pregnant women. The systematic implementation of the proposed relaxation techniques contributed in the reduction of perceived stress (mean change −3.23, 95% CI: −4.29 to −0.29) and increased the sense of control (mean change 1.99, 95% CI: 0.02—3.7). Conclusion: The findings suggest beneficial effects of relaxation on reducing perceived stress as well as increment of sense of control in pregnant women. The results of this study support the claim that training in the proposed relaxation techniques may constitute an ideal, nonpharmaceutical, intervention that can promote well-being, at least during pregnancy. Longer studies will be necessary in the future, in order to examine the long-term effects of relaxation techniques. © 2014 Elsevier Ltd. All rights reserved.
Corresponding author at: Soranou Ephessiou Str. 4, GR-115-27 Athens, Greece. Tel.: +30 210 6597644; fax: +30 210 6597644. E-mail address: [email protected] (C. Darviri). These authors contributed equally and share last authorship.
0965-2299/$ — see front matter © 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ctim.2014.01.006
204
Introduction Pregnancy for many women it may be particularly stressful, significantly affecting their quality of life, especially for the primigravida. Many factors have been associated with prenatal psychological stress. They generally include the sum of all arising demands, changes in body fitness, the shifting roles and the differentiation in interpersonal relationships with significant others.1 Although, these changes may be stressful, pregnancy is something most women choose to experience and as a result, birth is a welcome event usually accompanied by great enthusiasm. Moreover, in contrast to other stressful life events, pregnancy is a finite event with a specific duration and outcome which, in most cases, can be predicted with great accuracy.2 A significant number of studies in animals and humans have demonstrated that high maternal stress and mood disorders during pregnancy are associated with a variety of negative repercussions for both mother and foetus, including premature birth,3—7 higher risk of developing hypertension and preeclampsia,8 problems in lactation9 and increased vulnerability to mood disorders and autoimmune phenomena during the postpartum period.9 Furthermore, maternal stress has been shown to adversely affect the foetus’ development in the womb (brain development and neurological damage)10—15 and it is associated with the low birth weight of the infant, the small head circumference, the low Apgar score16—18 and with increased risk of miscarriage or stillbirth.19 It has also been demonstrated that maternal stress may later affect the relationship between mother and child as well as the child’s adult life (autism, schizophrenia, ADHD)20—25 and predisposes a difficult childbirth and the emergence of depressive symptoms during the postpartum period.26—29 In physiology, increased levels of stress, anxiety and depressed mood have been shown to correlate with changes in various parameters mainly through the activation of hypothalamic—pituitary—adrenal axis (HPA axis) and the sympathetic nervous system.30 Various hormones, including corticotropin hormone (CRH), adrenocorticotropic hormone (ACTH), cortisol and norepinephrine (NE), are released in large quantities in the bloodstream.30 How all these changes in physiological parameters may lead to adverse effects is not yet well understood. Furthermore, although research has focused more on the correlation between stress and possible complications in pregnancy, there is no significant bibliography demonstrating the usefulness of specific relaxation techniques in reducing stress during this period.31 It is anticipated that the systematic implementation of relaxation techniques may lead to important positive physiological changes which include a decrease in metabolism, oxygen consumption, cardiac function, blood pressure, breathing rate, and brain activity as well as to a significant increase in well-being.32 In the last decade, complementary-alternative treatments have been largely incorporated in routine medical practices, including pregnancy.33 Studies on the effect of relaxation techniques during pregnancy reveal a variety of significant effects.31 The systematic implementation of relaxation techniques has shown to reduce the likelihood of premature or delayed birth, to increase the infant’s body weight and to reduce the likelihood of a Caesarean
C. Tragea et al. section and birth interventions.34—38 Relaxation techniques that have been studied in pregnancy include most frequently yoga, hypnotherapy (guided hypnotherapy imagery script) and meditation with promising results in the reduction of stress, perceived stress, the reduction of various physical symptoms and the promotion of good mood and well-being.37—40 Applied relaxation, which is a combination of diaphragmatic breathing and muscle relaxation, has also been found to reduce anxiety and perceived stress and have other positive effects.34—36,41,42 There are limited data to support the efficacy of applied relaxation for stress reduction during pregnancy. This is the first study in Greek women, to examine change in perceived stress, anxiety and sense of control following a six week programme of a combination of two relaxation techniques (diaphragmatic breathing and progressive muscle relaxation). The purpose of this study was to evaluate the effects of applied relaxation (combination of diaphragmatic breathing and muscle relaxation) in reducing anxiety and stress in pregnant women during their second trimester, as well as raising their sense of control. By implementing this low cost non-pharmaceutical randomized controlled intervention, we expected a significant reduction in state-trait anxiety levels and in perceived stress and an increase in the sense of health control by participants in the intervention group compared with the control group. Also we expected to see a difference in some lifestyle factors associated with stress.
Method The design of the study was a randomized control trial (RCT), which included an experimental trial before and after the intervention (prospective pre-test—post-test experimental design). The main study hypotheses were as follows: (1) state anxiety scores of the experimental group will be significantly reduced compared with the control groups scores, after applied relaxation training, (2) trait anxiety scores of the experimental group will be significantly reduced compared with the control group’s scores, after applied relaxation training, (3) perceived stress scores of the experimental group will be significantly reduced compared with the control group’s scores, after applied relaxation training, and (4) sense of control will be significantly increased compared with the control group’s scores, after applied relaxation training. The sample group for this study was a convenience sample. It consisted of pregnant women invited from private obstetricians practice and the Attikon maternity hospital in Attica, more specifically from the department of Foetal Medicine. The study was accredited from the medical school of Athens and the Scientific Committee of Attikon hospital (registration number: UoAMedPR-4716-180211-23). The admission criteria for the study required that participants were primigravida, in the second trimester of their pregnancy (14—28 weeks gestation), aged 20—40 years, with singleton pregnancy and no prior history of medical complications. Also, they were required to understand Greek. Exclusion criteria were the therapeutic use of drugs for
Effects of a stress management programme medical complications and the systematic use of other relaxation techniques. The criterion for selection and monitoring of pregnant women in their second trimester was chosen because during the second trimester, pregnant women have organized the programme of maternity care they will follow. Additionally, there is enough time to implement a stress management programme (such as the specific which lasts 6 weeks), as well as enough time to evaluate the results before the major event of birth in another context (longer follow up) after birth. The study took place in Attica, from October 2010 to June 2011.The women, who participated in the study, after being informed about the purpose and process of the study, provided written consent prior to their participation in the study. The women who agreed to participate in the study were divided into two groups by randomization. The method of randomization used, resulted from an online random number generator based on cosmic rays (random.org). The two groups were the intervention group, which would implement the proposed relaxation techniques and the control group which would be given general guidelines in order to improve their lives and reduce stress. Both groups initially received a brief lecture on stress and the effectiveness of management techniques, while they were also given a brochure on stress which included the meaning of stress, its pathophysiology, the factors causally associated with it and the major consequences associated with pregnancy. They were also given brochures on the benefits of diet and exercise tailored for that specific period. In our study two types of interventions, diaphragmatic breathings and progressive muscle relaxation were used and were accounted for the observed effects. The education and counselling regarding healthy diet and regular exercise that both groups received in the beginning of the study consists a common practice in Greek maternity hospitals that is given to all pregnant women. The nutrition brochure included the fundamentals of the Mediterranean diet. These include regular small meals ensuring variety within the diet, with particular attention to breakfast. They were also given instructions on the frequency, quantity and type of food for each meal, with the aim of maintaining a stable body weight and receiving all essential nutrients tailored to the specific period of pregnancy they are going through. The brochure on exercise highlighted the importance of exercise and the fact that it can act effectively just like any other therapeutic agent. It included the main benefits of exercise, especially during pregnancy. A course of moderate exercise was recommended as well as activities generally recognized as safe during this period (walking, swimming, running and aerobics). The intervention group was given an audio cd (lasting 20 min with clear instructions) with the proposed relaxation techniques (diaphragmatic breathing exercises in combination with muscle relaxation exercises). This group was asked to apply the techniques as often as possible (1—2 times a day for 6 weeks). Relaxation through diaphragmatic breathing is accomplished by taking deep diaphragmatic breaths followed by
205 prolonged slow exhalation. The second part of the audio cd included progressive muscle relaxation (PMR) exercises. The process of PMR is designed to teach the way to relax body muscles through a two-step process. Initially, active contraction is applied to specific muscle groups. Then, the intensity stops and each individual turn her attention to observe how muscles relax after prolonged contraction. Moreover, they were given a diary to record and control the frequency of relaxation techniques, which they were asked to fill out each time they applied the relaxation techniques. Additionally, the intervention group was given a brochure containing instructions about the importance of adopting a healthy lifestyle through a routine and proper time management. Particular emphasis was placed on the importance of maintaining a programme in everyday life and time management, as well as adopting a routine which seems to have much effect in reducing stress and promoting good health. Monitoring the pregnant women was performed weekly by telephone communication or a personal meeting, when deemed necessary, throughout the duration of the programme, where the frequency of relaxation techniques was tested and the benefits of the techniques and any queries were discussed. For the sake of similarity between the two groups, there was a telephone communication once every week with the control group too. At the end of the six week programme, women in the control group were awarded by introducing the techniques accompanied with the relaxation cd. Data for the study were collected through self-report questionnaires that included the measurement of stress, anxiety and sense of control. Furthermore, a questionnaire was given, developed by the research team, measuring health and lifestyle — socio-demographic data and a few questions about the exact gestational age and the main concerns of pregnant women during that period. The questionnaires were handed out to the participants on the first meeting, as well as at the beginning of the programme and at the end after 6 weeks in both groups (intervention and control). Those participants who did not return completed questionnaires at the first meeting were called to return them after a period of three (3) days. The questionnaires used in this study were the following. The Perceived Stress Scale (PSS) is the most commonly used psychological tool for measuring the perception of stress. It includes a 5-point Likert-type scale ranging from never to very often. It is a tool for measuring the degree to which the situations in a person’s life are assessed as stressful. The questions are designed to demonstrate how unpredictable and uncontrollable the respondent considers the situations she experiences. The scale includes a series of 14 questions measuring perceived stress. The content of the questions is relative to the feelings and thoughts of the individual during the last month.43 A translated and validated version of PSS in Greek was utilized.44 The questionnaire of measurement of everyday life, lifestyle and health includes questions relating to the characteristics of sleep, eating habits, medical history, medical history, general health status, exercise and a list of various stress symptoms. This list does not constitute a psychometric tool. It consists of various symptoms associated in some degree with stress.
206
C. Tragea et al. 130 people were assessed
58 Not eligible 8 Declined to participate
85 Eligible
50 excluded 85 Randomized
44 Randomized to the Intervention
41 Randomized to the control group
6 weeks follow up
6 weeks follow up
4 Unable to contact
5 Unable to contact
3 Discontinued (for personal reasons)
4 Discontinued for personal reasons
1 stillbirth
3 Had medical complications
1 death of husband
12 Total
3 Had medical complications 1 Gave birth premature 13 Total 29 Included in primary analyses 31 Included in primary
60 participants
Figure 1
Flowchart of participants.
The Health Locus of Control, on a theoretical level, evaluates how people treat their health, as a result of their own behaviour (internal control centre for health), or as a result of other factors such as luck (external control centre for health). This is a self-report questionnaire, of 18 sentences. Respondents choose, on a scale of six, the degree to which each sentence represents them. Three subscales are calculated. The first measures the degree to which an individual feels they have control over their health (HLC1). The second measures the degree to which an individual feels that others are in control of their health (HLC2). The third subscale measures the degree to which an individual feels their health is a matter of luck (HLC3).45 A translated and criteria-cued version of HLC in Greek was utilized. The State-Trait Anxiety. The purpose of this questionnaire is to measure anxiety. In this scale, anxiety as a personality trait is differentiated from anxiety caused by
external factors. It consists of 40 topics, of which the first 20 refer to the respondent’s emotional state at the time of completing the questionnaire (i.e. transient stress as a result of the present situation) while the remaining 20 refer to the individual’s emotional state in general (i.e. constant stress as a personality trait). The evaluation was conducted according to the representativeness of the content of proposals for each respondent, on the basis of a 4-point Likert-type scale ranging from never to very much.46,47 The researchers used a version of STAI, translated and adapted into Greek. For the description of the qualitative features of the present study, absolute (number of observations) and relative (percentage) frequencies were used. The description of quantitative characteristics was based on the calculation of the median and the interquartile range — IQR. The statistical evaluation of the relationship between the group
Demographic and baseline characteristics of the participants at pre-intervention.
Age (years; median (IQR)) Gestational age (weeks; median (IQR)) Married (n, %) Planned pregnancy (n, %) Live in Greece (n, %) Education (years; median (IQR)) Education (n, %)
Intervention group (N = 31)
Control group (N = 29)
32.0 (28.0, 35.0) 17.0 (14.0, 21.0)
32.0 (29.0, 35.0) 17.0 (14.0, 20.0) 26 20 28 16.0
25 24 30 16.0
(80.6) (77.4) (96.8) (14.16)
(89.7) (69.0) (96.6) (14, 16)
Elementary High school Greater than diploma Income (in euros; n, %)
0 (0) 8 (25.8) 23 (74.2)
1 (3.4) 7 (24.1) 21 (72.3)
Low 1500
Intervention group (N = 31)
Control group (N = 29)
PSS-14 score (median (IQR)) A-state score (median (IQR))
23.0 (18.0, 27.0) 38.0 (35.0, 42.0)
22.0 (16.0, 27.0) 40.0 (30.0, 52.0)
A-trait score (median (IQR)) HLC1 score (median (IQR)) HLC2 score (median (IQR)) HLC3 score (IQR)
37.0 26.0 22.0 12.0
36.0 26.0 19.0 15.0
Satisfied with sleep quality (n, %) Smoking (n, %) Repressed anger (n, %) Sleep disorders (n, %) Complete control of things (n, %) Frequently Very frequently Lack of attention/inability to concentrate (n, %) Reduced memory capacity (n, %) Reported statement of stress (n, %) Tension headaches (n, %)
(31.0, (23.0, (18.0, (11.0,
42.0) 29.0) 26.0) 19.0)
(30.0, (24.0, (16.0, (11.0,
21 (67.7)
21 (72.4)
2 5 12 11
4 11 13 10
(6.7) (16.1) (38.7) (35.5)1 (3.2)
45.0) 29.0) 22.0) 20.0)
Effects of a stress management programme
Table 1
(13.8) (37.9) (44.8) (34.5)3 (10.3)
7 (22.6)
7 (24.1)
8 (25.8)
9 (31.0)
22 (71.0)
22 (75.9)
11 (35.5)
9 (31.0)
Interquartile range — IQR. PSS, Perceived Stress Scale; MHLC, Multidisciplinary Health Locus of Control (1, internal; 2, external; 3, chance); STAI, Anxiety; State-Trait Anxiety Inventory (S, state; T, trait).
207
208 (intervention—control) and the qualitative characteristics was based on Pearson’s Chi-squared (2 ) test. The statistical evaluation of the quantitative characteristics (by means of change: after minus before) according to group was conducted by applying the analysis of covariance (ANCOVA), after adjusting for baseline scores. Correlation between interval measures was calculated with Spearman’s rho coefficient. Level of significance was p 0.05. For the analysis we used the SPSS 17.0 software.
Results A flow-chart of the study appears in Fig. 1. A total of 142 women were initially evaluated during the aforementioned time period (October 2010 to June 2011). Of these, 52 individuals did not meet the entry criteria for the study, 10 were in the first trimester of their pregnancy, 13 were in the third trimester of their pregnancy, 18 were multiparous, 2 did not understand Greek, 5 systematically received medication for medical complications, 2 were carrying twins and 2 systematically applied other relaxation techniques. Of the remaining 90 participants, 5 refused to participate in the study for personal reasons. The number of pregnant women who agreed to participate in the study amounted to 85 individuals. Of the 44 individuals allocated in the intervention group, 3 were excluded because they experienced complications in their pregnancy and required systematic medication, 1 woman gave birth prematurely, 1 woman terminated pregnancy for medical reasons, 1 woman experienced an acutely stressful event (death of partner), 3 refused to continue the management programme and were excluded, while for another 4 phone communication was impossible. Of the 41 individuals allocated in the control group, 3 women had medical complications, 4 discontinued their participation for personal reasons while for the other 5 phone communication was impossible. As noted, both groups had similar dropouts. Subtle discrepancy was noted for 3 women with medical complications in the control group and 6 from the intervention group (3 with medical complications and 3 excluded due to stress, premature birth and stillbirth). The number of pregnant women who completed the stress management programme was, ultimately, 60 people. The median age (MIP) for women was 32 (29.35) years. The median pregnancy week on entry was week 17 (14.21). 85% of the participants were married and 73.3% stated that their pregnancy was planned. Also, 96.7% stated they had spent most their life in Greece. 73.3% had completed higher education while the majority reported a moderate socioeconomic level (n = 35) (Table 1). Baseline characteristics of the participants at preintervention are also showed in Table 1 and no significant differences were noted. At the end of the study, it was observed that the percentage of pregnant women who reported being happy with their sleep was higher amongst individuals in the intervention group (87.1%) in comparison with the control group (62.1%), (p-value = 0.025). Also, significantly (p < 0.05) more women in the control group reported continuation of smoking (13.8%), a sense of repression and
C. Tragea et al. anger (20.7%), sleep disorders (41.4%) in comparison with those in the intervention group (0%, 0% and 16.1%, respectively). Finally, it was observed that pregnant women in the intervention group (74.2%) reported a higher percentage of a frequent sense of complete control of things during the final month, in comparison with those in the control group (41.4%), at the end of the study (p-value = 0.018) (Table 2). At the end of the study, significant differences were observed in the scores of the questionnaires examined (Table 3). PSS-14 scores were significantly reduced (mean change −3.23, 95% CI: −4.29 to −0.29) compared to controls. Also, significant improvement for internal health locus of control (mean change 1.99, 95% CI: 0.02—3.7) was noted for women in the intervention group. No other significant difference was noted for the other variables (Table 3). Most women in the intervention group practiced the relaxation techniques for at least once per day (42 times) out of the 84 proposed during the 6 week period, while ten of them practiced the relaxation CD 84 times in total during the six week programme. Correlations between compliance with the stress management technique and mean outcome changes revealed that compliance was significant only for internal health locus of control change (Spearman’s rho = 0.411, p value 0.02).
Discussion The present study examined the effectiveness of two relaxation techniques (progressive muscle relaxation and diaphragmatic breathing) during pregnancy. The results of the study demonstrate significant benefits from the use of the techniques in the psychological state of the pregnant women. The systematic implementation of the proposed relaxation techniques contributed in the reduction of perceived stress and anxiety and increased the sense of internal control. Also we found changes in many lifestyle factors associated with stress. Our results replicate previous studies both for lifestyle factors (such as sleep, quality of life, etc.) and psychosocial ones (such as stress, health locus of control, anger, etc.). In specific, few studies confirm our positive results on quality of sleep, less sleep disorders, less feeling of repression and anger which may reflect the benefit of lowering stress and anxiety levels.36,48 Also, one study, using a big sample of 110 pregnant women, significantly reduced anxiety and perceived stress.34 PMR along with other techniques can also reduce stress levels during pregnancy, baseline heart rate and blood pressure, which is related with stress levels.36,41 More interestingly, stress management can beneficially affect pregnancy-related outcomes such as low birth weight, frequency of Caesarean sections and length of pregnancy.35,42 Direct comparisons of our study with the aforementioned are not possible due to different tool utilization for measuring psychosocial variables. We are aware that the results found in our study may have been partially differentiated in the intervention group due to the existence of a trainer—consultant. It is also possible that the more frequent phone and/or personal contact with the trainer in the experimental group may have helped to reduce stress and anxiety, apart from the implementation
Effects of a stress management programme Table 2
Comparisons of lifestyle characteristics of the participants at post intervention.
Satisfied with sleep quality (n, %) Smoking (n, %) Repressed anger (n, %) Sleep disorders (n, %) Complete control of things (n, %) Lack of attention/inability to concentrate (n, %) Impaired memory capacity (n, %) Reported statement of stress (n, %) Tension headaches (n, %) *
209
Intervention group (N = 31)
Control group (N = 29)
27 (87.1) 0 0 5 (16.1) 23 (74.2%) 8 (25.8) 9 (29.0) 17 (54.8) 4 (12.9)
18 (62.1) 4 (13.8) 6 (20.7) 12 (41.4) 12 (41.4%) 3 (10.3) 5 (17.2) 17 (58.6) 6 (20.7)
0.025* 0.046* 0.028* 0.023* 0.018* 0.093 0.097 0.768 0.14
Level of significance p < 0.05.
of the techniques. In order to minimize we had organized weekly telephone communication with women of both groups during which concerns and queries were discussed. Thus sense of offer was fostered in all women. Moreover it was not possible to monitor and control adherence to the stress management programme by the participants, besides the diary given for recording the frequency of technique implementation. Most of the participants of this study were married, with a desired pregnancy and had had several years of education; therefore results may not be generalized and may differ in unmarried women, with an unintended pregnancy and lower level of education. The profile of the participants in our study (desired pregnancy, well educated) might have influenced their adherence to and expectation by the proposed relaxation techniques at home in order to ensure a healthy pregnancy, as another study has shown.49 Another limitation is the per protocol analysis, although similar dropouts were noted in both groups. Moreover one more limitation of the study involves the measurement of pregnancy specific stress by using questionnaires designed to measure general stress. Many research groups distinguish pregnancy specific stress,50,51 proposing specific tool for measuring stress (such as the Prenatal Distress Questionnaire, the Pregnancy Related Anxiety Questionnaire and the Pregnancy Experience Scale) as more powerful and reliable
Table 3
compared with others not related to pregnancy, especially in predicting premature birth.2,51—53 The results suggest that stress management during pregnancy resulted in a significantly decline in anxiety and perceived stress. Also a positive affect in the sense of control was also found. Despite the limitations, our results support that training in the proposed simple, cheap and easily implemented relaxation techniques may constitute an ideal, non-pharmaceutical, intervention that can promote well-being, at least during mid-pregnancy. This pilot study suggested that stress management during pregnancy is feasible and preliminary evidence from this sample supports its potential efficacy in this areas, particularly if started early in the pregnancy. Lengthier follow-up studies will be necessary in the future, in order to examine the long-term effects of relaxation techniques by using pregnancy specific stress tools. Based on these preliminary results and given the poor literature in the subject, we strongly support that health professionals — and more specifically midwifes and obstetricians — should be educated in relaxation techniques, and in turn to encourage women to learn and implement relaxation, in order to study further the possible benefits in pregnant mental health, in pregnancy outcomes and childrearing.
Mean changes (95% confidence interval) in scores (end—beginning of study) according to each group.
PSS-14 score A-state score A-trait score HLC1 score HLC2 score HLC3 score
Intervention group
Control group
Mean (95% CI)
Mean (95% CI)
−3.68 −3.53 −3.84 1.71 −0.32 −0.23
−0.45 −2.03 −1.55 −0.28 1.48 −0.55
[±1.80] [±2.80] [±1.43] [±0.98] [±1.43] [±1.25]
[±1.80] [±2.88] [±2.46] [±1.64] [±1.07] [±1.45]
Between group difference 95% CI
−3.23 [−4.29 to −0.29]* −1.5 [−2.7 to 1.7] −2.29 [−4.9 to 0.3] 1.99 [0.02 to 3.7]* 1.8 [−0.7 to 3.5] −0.32 [−2.12 to 2.01]
ANCOVA using baseline scores as covariate. PSS, Perceived Stress Scale; MHLC, Multidisciplinary Health Locus of Control (1, internal; 2, external; 3, chance); STAI, Anxiety; State-Trait Anxiety Inventory (S = state, T = trait). * 95% confidence interval (CI) that do not contain the value zero (p value
