Depressive symptoms, anxiety, and quality of life of Japanese women at initiation of ART treatment

Assisted reproductive technology (ART) treatment accounted for 6% of total births in 2017 and is increasing which places Japan among the top worldwide in number of treatments performed. Although ART treatment patients often experience heavy physical and psychological burden, few epidemiologic studies have been conducted in Japan. We examined mental health and health-related quality of life (QOL) among women at early stages of treatment. We recruited 513 women who have initiated ART treatment, either in-vitro fertilization or intracytoplasmic sperm injection, from four medical facilities in the Tokyo area and through web-based approaches. At baseline, we collected socio-demographic information and assessed depressive symptoms, anxiety, and QOL. Descriptive analyses were performed overall and stratified by factors such as age. Mild depressive symptoms or worse, assessed with Quick Inventory of Depressive Symptoms, were observed among 54% of participants. Mean score for State-Trait Anxiety Inventory was 52 with a standard deviation of 11 for the state, and 39% were categorized as high anxiety. QOL results, assessed with SF-12, showed the same negative tendency for social functioning and role (emotional), while general health and physical functioning were consistent with the national average. Young participants appeared to suffer mentally more than older participants (p < 0.01 for depressive symptoms). Our findings suggest that patients may be at high risk of depressive symptoms, high anxiety, and low QOL even from the early stages of ART treatment.

Despite Japan being number one in the world in terms of number of ART treatments, epidemiologic studies investigating physical and mental health of patients undergoing infertility treatment remain scarce. To our knowledge, there are only three studies published in international journals [18][19][20] , and no studies have exclusively focused on ART patients. To address this gap, we have initiated a prospective cohort study of women receiving ART treatments; in this report, we introduce the design and methodological features of this ongoing study, and describe the depressive symptoms, anxiety, and QOL, as well as socio-demographic characteristics of women who are in early stages of ART treatment.

Method
Study population and recruitment. Women at early stages of ART treatment were enrolled into our cohort study. The eligibility to participate in this study included patients who were either, (1) starting the ART treatment (IVF or ICSI) with no prior experiences, or (2) had started the treatment and retrieved oocytes once or twice at the time of recruitment. We employed two methods of recruitment for this study, one of which was through clinicians at medical facilities. Participants were recruited at four medical facilities in Tokyo and Kanagawa prefectures, which offer infertility treatment. The clinicians distributed a study envelope that consisted of a description of the study, consent form, and baseline survey to eligible patients at the initial counseling session. Those willing to participate filled out the consent form and survey and returned them using an enclosed envelop with the return address and postage. A total of 74 participants were recruited from the medical facilities. The second approach to recruitment comprised the use of a study website which was advertised through social networking services and media-related outlets. Those who visited the website and had interest in participating registered their mailing address, and a study envelope was mailed to them. A total of 439 from 550 registrations participated (80% participation rate). We included those with children as long as they had no experience with ART in prior pregnancies and excluded those who were already pregnant as a result of treatment. The total number of participants was 513, and informed consent to participate in this study was obtained from each participant. Reported content followed the STROBE checklist for cohort studies. This study was performed in accordance with the Japanese Ethical Guidelines for Medical and Health Research Involving Human Subjects, and an approval from ethics committee for this study was obtained at the National Center for Child Health and Development (No. 1993).
Follow-up. Participants are followed for 1 year during which time a series of six questionnaires are administered sequentially every 2 months in order to evaluate the trajectories of depressive symptoms, anxiety, and QOL as multiple cycles of ART treatment are pursued (i.e., oocyte retrievals and embryo transfers). The number of oocyte retrievals and embryo transfers were ascertained at the baseline survey and at follow-up. Participants are followed up to 1 year or until time of withdrawal or loss to follow-up, and are censored at the point of achieving pregnancy.

Main outcomes.
The main interests of this study are depressive symptoms, anxiety, and QOL among women who begin ART treatment. We used QIDS (the Quick Inventory of Depressive Symptomatology-Self report) to assess depressive symptoms, STAI (the State-Trait Anxiety Inventory) to assess anxiety, and SF-12 to assess QOL. Using the total score for QIDS, we created a five-category variable 21 : none, mild, moderate, severe, and very severe. Using the total scores for state and trait, each was categorized into 5 levels where 1 represented the lowest anxiety level and 5 was the highest; we defined categories 4 and 5 as high anxiety 22 . Based on the SF-12 responses, scores for 8 sub-scales were calculated: physical functioning, role (physical), bodily pain, general health, vitality, social functioning, role (emotional), and mental health. The raw scores were transformed into norm-based scores with the average of 50 and a standard deviation of 10, in which lower scores indicated signs of low QOL 23 .
Socio-demographic variables. We collected a range of socio-demographic information: age, educational attainment, employment status such as type of employment and company size, residing prefecture, and income of the women/patients and their partners. Type of employment was grouped into three categories: full-time, parttime, and not working. Part-time included those who worked part-time, were contractual, or self-employed. Not working included unemployed, housewife, and student. Education was classified into three levels: high school graduate or below; vocational school or 2-year college; and 4-year college or graduate school. We also obtained information on height and weight to calculate body mass index, durations for marriage and infertility treatment (e.g., intrauterine insemination), reasons for wanting a child, sources of emotional support, a sub-scale for partner relationship 24 , and method of financing for the treatment. We inquired about the age and sex of children if there were any and the method of their conception (i.e., naturally conceived or not).
Despite the increasing popularity of ART treatment, characteristics of patients remain unknown. To further understand socio-demographic characteristics of ART patients in relation to the general population, we utilized 2016 data from the Comprehensive Survey of Living Conditions, comprising a nationally representative sample of Japanese families conducted by the Japanese Ministry of Health, Labour and Welfare, for comparison 25 . From the national data, we extracted married women between the ages of 20 and 45 who do not have children (n = 7468).

Statistical analysis.
We performed descriptive analysis to examine the proportions across categories of QIDS and STAI and average scores for the SF-12 sub-scales among the participants. Additionally, we stratified QIDS and STAI categories and SF-12 sub-scales by number of oocyte retrievals, number of children, and age categories and assessed whether or not the pattern of depressive symptoms, anxiety, and QOL varied by these www.nature.com/scientificreports/ factors using chi-square tests and simple linear regression. We treated number of oocyte retrievals and age categories as ordinal in the simple linear regression analysis. A two-sided p value of less than 0.05 was considered statistically significant. All the analysis was done with STATA 13 (Stata Corp., College Station, TX, USA).

Results
In Table 1, we present socio-demographic information of study participants and of women sampled from a nationally-representative survey of the general population (married women with no children). Compared to the general population, we observed a larger proportion of women in their 30 s in this study where the mean age was 35 years (range 24-46 years). The proportion of full-time employment and working at larger companies was higher in this study than the general population (full-time employment: 48% vs. 37%; working at a company with ≥ 1000 employees: 27% vs. 19%), as was the proportion of highly educated women (65% vs. 27%). Participants were mainly from Tokyo and neighboring prefectures. Although income information was not available for the general population, we expect that study participants likely had higher earnings. Half of the participants noted that the reason for infertility was unknown. There were 28 participants (5%) who had 3 or more experiences with oocyte retrieval at baseline. Although they were not eligible to participate in this study, we retained them for comparison purposes.
In Table 2, we present levels of depressive symptoms, anxiety, and QOL among the participants. The mean score for the QIDS was 7.0 (SD 4.7), and the state and trait scores for the anxiety scale were 51.7 (SD 11.3) and 52.8 (SD 11.4), respectively. Categorization of the QIDS score showed that 54% of participants were already showing some signs of depressive symptoms. Anxiety also seemed to be high among the participants (39% for categories 4 and 5 for state). We also observed similar patterns with the QOL sub-scale scores. Norm-based scores showed that social functioning, role (emotional), and mental health aspects were low in particular, while physical functioning and general health were near the average.
In Table 3, we present depressive symptoms, anxiety, and QOL stratified by number of oocyte retrievals, number of children, and age categories. We did not observe a clear pattern with the exception of physical role functioning appearing lower among those with more oocyte retrievals. We observed those with one or more children seemed to be doing better mentally. Young participants (i.e., < 30.0) seemed to show higher prevalence of depressive symptoms and higher anxiety and lower QOL scores than older participants.

Discussion
In this report, we described the profile of a new longitudinal study aimed at addressing gaps in the current epidemiological literature on the mental health and well-being of women undergoing ART treatment in Japan. We provided a first report focused on depressive symptoms, anxiety, and QOL as well as socio-demographic profiles of women who were starting or had just started ART treatments. Results showed that indicators of mental health already appeared unfavorable at the early stages of ART treatment. Descriptive analysis of the sociodemographic characteristics showed that those who begin ART treatment tended to be in their thirties residing in Tokyo or adjacent prefectures, with high education and full-time employment. Although Japan leads the world in the number of ART treatments performed per year, epidemiologic studies on the mental health and QOL of patients remain scarce. To our knowledge, this study presents one of the first opportunities to demonstrate the mental health status of Japanese women struggling with infertility and starting ART treatment to have a child. High prevalence of poor mental health at early stages of treatment is concerning because high stress has been associated with discontinuation of infertility treatment and may have profound influences on daily lives such as sexual functioning 13,26 .
Aligned with previous findings from Japan and other countries, women who go through ART treatments appeared to be highly stressed, indicated by depressive symptoms, high anxiety, and low QOL 9 . One study conducted in Japan showed that 32% of patients undergoing infertility treatment showed depressive symptoms, based on the Center for Epidemiologic Studies-Depression Scale 27 . In another Japanese study comprising 125 patients receiving ART or other infertility treatments, the average scores of STAI were 46.5 (SD 9.5) for trait and 48.6 (SD 8.6) for state 28 . Matsubayashi et al. showed that scores based on the Hospital Anxiety and Depression Scale and Profile of Mood States were higher among infertility patients than pregnant women 18 . Our results were mostly consistent with these previous findings despite the differences in tools to assess mental health. A study by Ogawa et al., also showed elevated levels of anxiety and depression, but the prevalence appeared higher in older patients which was in contrast with our findings of higher prevalence among younger age groups 20 . To our knowledge, no previous studies have examined QOL among infertility treatment patients in Japan.
We showed in our study that women at early stages of the ART treatment were already showing signs of deteriorated mental health and QOL. Reasons for this tendency remain unknown and are to be examined, but one contributing factor may be the expectation of high treatment cost. The total treatment cost could easily reach up to a million Yen or more as they go through multiple cycles of treatment. Another factor may be related to time management needed to coordinate treatment and work issues. Because patients cannot know the schedule of treatment in advance, and frequent clinic visits are necessary, such time constraints could interfere with work. Japanese workers are known for their long work hours 29 , and the corporate culture may cause logistical difficulties for those working full-time. In our study, half of the women were working full-time.
Half of the participants noted that reasons for infertility were unknown. Thus, they do not know which treatment would be best for them to increase the chance of achieving pregnancy. This uncertainty regarding whether the selected treatment is the most effective for their condition may increase stress and anxiety levels. In Japan, the success rate of achieving pregnancy at each fertility clinic is often not available publicly to the patients, while in other countries such as the United States, government agencies publish summary reports 30  www.nature.com/scientificreports/ This lack of information may add to the unforeseeable nature of ART treatment, which is likely to be one of the main sources of stress for patients. Stratified by number of oocyte retrievals, number of children, and age, we observed certain patterns regarding depressive symptoms, anxiety, and QOL. For example, those with at least one child seemed to be better off mentally than those without. This observation may be influenced in part by the decision-making process of choosing to receive the treatment. Only those with suitable circumstances, such as financial capacity, may decide to receive infertility treatment to have another child. For others, the option may not even be available to them due to sub-optimal circumstances. In addition, for those who already have a child, treatment failures may not be as pressuring compared to those who desperately want a first child. High prevalence of depressive symptoms and low QOL among young participants may be explained by the general perception that age is a major contributing factor of infertility. The need for ART treatment may be perplexing and unpleasant to the young participants. Some may have serious health problems which could lead to infertility and also influence mental health negatively. In addition, because of the seniority system and employment structure in Japan, wages for young people tend to be low and may contribute to additional strain on mental health 31,32 .
Our results indicate the need for mental health care support for infertility treatment patients even at the early stages of the treatment. In other countries, intervention studies have been conducted to reduce distress among infertility treatment patients 33 . However, in Japan, such studies have been scarce despite the popularity of infertility treatment. In a study by Asazawa, it was reported that a partnership support program could reduce psychological distress among women 34 . The Japan Society for Reproductive Psychology has translated a publication by the European Society of Human Reproduction and Embryology, "Routine psychological care in infertility and medically assisted reproduction-A guide for fertility staff " into Japanese, and this translated literature is made available on their website 35 .
It is important to acknowledge certain limitations of this study. Because we could not employ random sampling, our results may not accurately reflect the circumstance of the broad-level population receiving infertility treatment in Japan. There are no national data on socio-demographics of the women who go through ART treatments. We may be overestimating the prevalence of depressive symptoms, high anxiety, and low QOL because it is possible that those who wanted to address the difficulties of treatment may have had a higher likelihood of participation. However, the opposite is also conceivable in which those who were not doing well mentally might have had a tendency not to participate. Although we used validated scales such as QIDS, STAI, SF-12, www.nature.com/scientificreports/ they were not specifically designed to assess special populations such as those who go through ART treatment. In European countries, scales such as Copenhagen Multi-Centre Psychosocial Infertility Fertility Problem Stress Scales (COMPI-FPSS) and SCREENIVF have been developed and used with infertility treatment patients 36,37 , but these were not available in Japanese except for FertiQOL 38,39 . Scales designed for this special population may be necessary. Finally, it is important to note that this descriptive report is based on baseline data taken from a longitudinal survey; thus, it does not reflect the temporal sequence of events. For instance, we do not have access to the participants' mental health information prior to their initiation of infertility treatment. However, epidemiologic studies on patients who undergo ART treatment are generally lacking in Japan. As such, we believe that descriptive data presented in the current report is important as basis to understanding the problem and informing the direction of future initiatives. Using data from the follow-up surveys, we plan to examine the trajectory of mental health and QOL and potential risk and protective factors that could influence the trajectory 40 .

Conclusion
Despite the high demand of ART treatments in Japan, studies examining potential negative effects of the infertility treatment on patients' body and mind remain scarce. In addition to epidemiologic studies to confirm our findings, intervention studies to alleviate the stress of the ART treatment among patients may be needed.