303 JOURNAL OF SCIENCE, Hue University, N 0 61, 2010 EXPLORING POSTNATAL DEPRESSION IN THUA THIEN HUE PROVINCE, CENTRAL VIETNAM Linda Murray, Michael Dunne, Nigar Khawaja School of Public Health, Queensland University of Technology (QU) Cao Ngoc Thanh College of Medicine and Pharmacy, Hue University SUMMARY Introduction: Postnatal depression (PND) is an important public health issue due to its impact on maternal wellbeing, infant development, and family cohesion. The estimated prevalence of PND during the first 12 months post-partum ranges between10-20% worldwide. Whilst PND used to be considered a syndrome only occurring in western countries, there is now evidence that it occurs throughout the world, and often at higher rates in low and middle- income countries. To date, there has been little research into PND in South East Asia and only two community-based surveys in Vietnam, one in Ho Chi Minh City in 1999 and one in Hanoi and Ha Nam in 2009. This study will investigate health worker attitudes about risk and protective factors for PND among women in Thua Thien Hue province in central Vietnam. Methodology: In 2009, 23 health professionals participated in qualitative exploratory research of postnatal depression in Hue. This included two focus groups with 12 health professionals who completed a concept mapping process, and in-depth interviews with another 11 health professionals. Results: Many factors relating to postnatal depression were identified including socio-economic status, son preference, mother’s health, infant health, social support from family and the community, and health promoting behaviours. In-depth interviews highlighted community knowledge and attitudes surrounding PND such as traditional confinement practices and fear of experiencing stigma. Conclusion: The findings of this research will be used to plan a substantial community-based quantitative survey in order to establish prevalence of PND and surrounding social determinants in central Vietnam. 1. Introduction Postnatal depression (PND) is a significant public health issue which not only impacts maternal wellbeing, but also infant growth and development, and family cohesion. Prevalence of PND is estimated to be between 10-20% worldwide, although rates vary within and between countries. Before the 1980’s, it was assumed that PND only occurred in western countries. However once PND began to be measured internationally using standardised tools, it became apparent it was a genuine universal 304 disorder showing similar symptoms across cultures. Emerging evidence suggests the prevalence of PND may be higher in low and middle income countries than in high income countries, and carry considerable social and economic consequences. However less than 10% of low and middle income countries have data available. Socio-cultural factors are very influential in the aetiology and progression of perinatal mental health disorders. Social risk factors of PND identified in Hong Kong, China, Singapore and Thailand include unwanted pregnancy, intimate partner violence, relationships with extended family (particularly in-laws), maternal self esteem, infant temperament and socio-economic status. Son preference is cited as an influential factor of PND in China, Hong-Kong, Taiwan and Korea. In mainland China, mothers who gave birth to a girl are twice as likely to suffer from PND. A study of 252 postpartum women in Goa, India, also found that the risk of PND following disappointment about the infant’s gender had an odds ratio of 3.3 (p = 0.002). In Vietnam, son preference has not been studied in relation to PND. However Pham et al (2008) state that the national reproductive health policy can create pressure to have a son within two births. Studies of perinatal mental health have been conducted in both the north and south of Vietnam. In a sample of 506 women in HCMC, 33% of women were found to have PND, and 19% of participants acknowledged thinking about suicide. In the north of Vietnam, a study of 364 postpartum women from Hanoi and Ha Nam province found that 29.9% had a common mental disorders (CMD) and that CMD were more common in rural areas. Other studies have found that 20% of mothers of one year olds had anxiety or depression in Vietnam, and that between 8 and 16.9% of recorded perinatal deaths could be attributed to suicide. There is currently no data on PND from central Vietnam. The purpose of this research is to identify risk and protective factors that health professionals in Thua Thien Hue, Vietnam, perceived to be related to PND in order to plan for a larger epidemiological study of maternal mental health in the area. 2. Methodology Methodological Approach: The theoretical approach of this research project was community based participatory research (CBPR). CBPR aims to draw on multiple sources of expertise when deciding what health risks matter, what causes them and what can be done about them. In social epidemiology, CBPR is used to increase the rigor of descriptive research through identifying social determinants of health through collaboration with local stakeholders Study Site: This study was conducted in Thua Thien Hue province in Central Vietnam in collaboration with Hue University of Medicine and Pharmacy. Participants: For the focus groups, a purposive sample of twelve maternal and child health professionals were recruited from three commune health centres within 10km of Hue city, and one Provincial hospital. Nine midwives with 2-32 years 305 experience, and three doctors with 15-18 years experience participated in 2 two-hour sessions. Another eleven experts in maternal health and mental health such as Obstetricians and Psychiatrists participated in in-depth interviews. Ethics: This study has ethical approval from both Queensland University of Technology and Hue University of Medicine and Pharmacy. Methods: In-depth interviews lasted 40-60 minutes and were recorder on an MP3 player with the interviewee’s permission. If the interviewer did not wish to be recorded the researcher took detailed notes. Interviews were either conducted in English or translated simultaneously. The interview transcripts were then analysed into themes using NVivo 8 software. For the focus groups, a concept mapping process developed by Trochim (1989) was used. Trochim’s concept mapping is a structured group process which both describes concepts as well as quantifying the strength of conceptual relationships. The focus groups were conducted by two Vietnamese facilitators, who translated materials into English at the end of the workshops. At the end of the concept mapping process, the results are visually represented as concept ‘maps’ using ‘Concept Systems’ software. Trochim concept mapping uses the following steps: Brainstorming: Participants were given a definition of postnatal depression and it’s common symptoms, and asked to think of any factors they thought might be related to it, whether positive or negative. Sorting: Each participant then received a packet of the statements printed on cards, and were asked to sort them into piles ‘in a way that made sense to them.’ The rules were that each card could only be used once, and to make more than one pile. Once they had finished the individual sorted piles were collected and entered into Concept Systems software for analysis. Rating: Each participant was then asked to rate each statement from 1 ‘least likely’ to 5 ‘most likely’ according to the following headings: ‘likely to cause PND’ and ‘likely to protect against PND.’ This information was also analysed with Concept Systems software. Verification: Once the information was analysed, the groups then met again to look at the concept maps and decide if they accurately reflected their previous work. 3. Results 3.1. Brainstorming: Participants brainstormed 46 statements to include in the sorting and rating activities. A list of the statements is outlined in figure one. 3.2. Sorting: From the sorting exercise, a concept map was produced (see figure 306 two). Each of the 46 statements is located as a point on the map. Statements that were sorted together more frequently appear closer to each other on the map and represent similar themes than statements more distant from them. The statements fitted naturally into eight thematic clusters, which are enclosed by the polygons on the map. 3.3. Rating: The next two maps show how the eight clusters were rated by participants as ‘most likely to cause PND,’ and ‘most likely to protect against PND.’ The rating map depicts the relative average rating on a 5 point likert scale (where 1 = least likely, 5 = most likely) in relation to being likely to cause PND. The number of layers represents how strongly participants rated a statement as being likely to cause PND. The more layers there are, the more participants rated statements in this cluster as likely to cause PND. The layers represent a double averaging – across all participants and all of the statements in each cluster. Figure three clearly shows that the clusters ‘negative stressors,’ and ‘mother’s emotions and worry’ were rated as highly likely to be risk factors for PND with average ratings of 3.73 to 4.24. ‘Mother’s health’ and ‘baby’s health’ were also considered to be risk factors with average ratings of 3.23 to 3.75, but were not rated as highly as the previous two clusters. ‘Economics’ had an average rating of 2.73 to 3.23, which means it was not considered as influential as other social and health factors in causing PND. Figure four represents statements participants perceived as most likely to protect against PND. Economics was moderately rated as a protective statement (average rating of 2.72 to 3.08), although it should be noted participants put positive as well as negative statements about economics in this category. Interestingly, ‘family aspects’ and ‘society and friendship’ (average 3.82 to 4.91) were rated more likely to be protective than factors explicitly categorised and labelled in focus groups as ‘protective factors.’ The statements in the ‘protective factors’ cluster were mainly about individual health behaviours such as exercise and rest, whereas ‘family aspects’ and ‘society and friendship’ all describe personal relationships and connections. 3.4. Interpretation of Maps In the second focus group, participants were shown the maps and decided if they considered the clusters generated to be meaningful. Both groups agreed that statement 16 ‘sleeplessness because of the baby’ should be moved to the cluster ‘Mother’s Health’. The group of 6 midwives thought that husbands needed a category of their own, apart from other family aspect, however it was decided overall that this wouldn’t be changed. In general, both groups thought the maps represented their ideas accurately. 3.5. In – depth interviews In-depth interviews provided more depth to many of the themes discussed in the focus group, including themes of son preference and also stigma about accessing mental 307 health services. The quote below explains women’s reluctance to access health services: “They don’t come to the hospital except for the serious case, the reason is they don’t want to meet with the psychiatrist because sometimes the depression in women makes them not want to take care of the baby and if they go to the psychiatrist, maybe they, the psychiatrist will say they are lazy or that they don’t love the baby.” One of the interviews provided the following vignette about a woman suffering from possible PND in relation to son preference: Interviewee: She has a patient, a woman who is a teacher, she had the first baby but unfortunately she is a girl, so after delivery she calls this postnatal depression because the woman don’t like to contact with other people, she just lies, the back of hers to the baby and doesn’t want to eat, she didn’t want to eat and didn’t want to talk to other people, because of the gender… Because when she was pregnant and she took an ultrasound and the doctor said that this was a boy and the husband was very hopeful, but after delivery the baby was a girl so the husband was a little bit disappointed and it makes the woman feel bad. 4. Discussion Overall, the concept mapping exercise allowed participants to identify risk and protective factors of PND both individually and as a group. Participants naturally sorted the statements into eight clusters, some which were previously in the literature on PND such as maternal health, infant health, social relationships, economics and family relationships. Interestingly, some of the statements that emerged are not previously identified in the literature such as ‘good atmosphere,’ ‘worry about beauty and physical appearance,’ ‘enough rest, sleeping late,’ ‘the grandparents not accepting the child and sending it to an orphanage’ and statements regarding son preference. The fact infant gender was mentioned in five separate statements highlights its importance. This is congruent in other literature from Asia but is yet to be explored in - depth in. Other statements in the ‘emotion and psychology’ cluster such as ‘physical appearance,’ ‘low self esteem within the community,’ and ‘gender of the baby the husband didn’t want’ have also not been explored in relation to each other before. Mother’s health, baby’s health and economics are routinely mentioned in the literature about PND from other countries, but were not considered as important as categories including social support. Rating social factors more highly than economics as a risk factor for PND is also consistent with the literature on risk factors for PND in developing countries. Vietnamese mothers have very high levels of social support, with one study showing 85% of mothers received support from two or more sources. Strong social support positively affects maternal coping, child nutritional status and cognitive development. The benefits of supportive nurturing family relationships can override even severe economic adversity in protecting against common perintatal mental 308 disorders. In-depth interviews added depth to the themes mentioned above. They also highlighted that many women suffering mental health issues in Hue were unlikely to seek treatment due to little knowledge about PND or fear of experiencing stigma. It is internationally recognised that large proportions of people suffering from common mental disorders in low and middle income countries receive no treatment due to underuse of, and poor access to health services and stigma In a study of maternal mental health in northern Vietnam, none of the women diagnosed with a CMD had ever received mental health care. Limitations of this method include the fact it requires resources that may not be easy to attain such as proprietary software One major limitation of the method’s accessibility is that Concept Systems is yet to be available with fonts for other languages. Also, as this is a qualitative method, the results are not generalisable, but are only representative of the ideas of the group that participated. Hence the results are useful for planning further studies but not identifying social determinants at a population level. 5. Conclusion Overall, this exploratory research was useful in identifying health professional’s perspectives about the social determinants likely to influence PND in Hue. The concepts resulting from this research will be used to inform a larger quantitative epidemiological study of social determinants of PND in Hue in 2010. REFERENCES 1. Beck, C. T., & Driscoll, J. W. Postpartum mood and anxiety disorders: A clinician's guide. Massachusetts: Jones and Bartlett Publishers, (2006). 2. Chien, L., Tai, C., Ko, Y., Huang, C., & Sheu, S Adherence to "doing-the-month" practices is associated with fewer physical and depressive symptoms among postpartum women in Taiwan. Research in Nursing & Health, 29(5), (2006), 374-383. 3. Das, J., Do, Q. T., Friedman, J., & McKenzie, D Mental health patterns and consequences: Results from survey data in five developing countries. The World Bank Economic Review, (2008), 1-25. 4. De Silva, M. J., Huttly, S. R., Harpham, T., & Kenward, M. G. Social capital and mental health: A comparative analysis of four low income countries. Social Science & Medicine, 64(1), (2007), 5-20. 5. Fisher, J. R. W., Tran, T., Buoi, L. T., Rosenthal, D., Kriitmaa, K., & Tuan, T Common 309 perinatal mental disorders in women in the north of Vietnam: Community prevalence and interaction with health care use. Bulletin of the World Health Organsation, In Press, (2010). 6. Gupta, M. D., Zenghua, J., Bohua, X. Z. L., & Chung, B. H. W. (2003). Why is son preference so persistent in east and south asia? A cross-country study of China, India and the Republic of Korea. World Bank Policy Research Working Paper no.2942. The World Bank. Retrieved 29 November 2009, from http://ssrn.com/abstract=636304 7. Harpham, T., & Tuan, T. . From research evidence to policy: mental health care in Vietnam. Bulletin of the World Health Organization, 84(8), (2006), 664-668. 8. Kane, M., & Trochim, W. M. K Concept Mapping for Planning and Evaluation. Thousand Oaks: SAGE, (2007). 9. Kleinman, A Global mental health: a failure of humanity. The Lancet, 374, (9690), (2009), 603-604 10. Wong, J., & Fisher, J The role of traditional confinement practices in determining postpartum depression in women in Chinese cultures: A systematic review of the English language evidence. Journal of Affective Disorders, 116(3), (2009), 161-169. Figure One Cluster Statements Economics 2 Worry about losing employment (applies to professionals) 11 Lost opportunities for promotion (e.g after third child) 1 Economic difficulty 9 Loss of income post delivery 10 Sufficient financial resources 4 Less time for social activities Family Aspects 25 Informational support from relatives and family 44 Happy/harmonious family 21 Supportive husband 22 Gender of the baby is the one they desired Society and Friendship 26 Support from the reproductive health program 40 Good atmosphere 43 Help of the neighbourhood 46 Support from mother’s friends Mother’s emotions 34 Single mother 310 and worry 31 Husband very concerned about the gender of the baby 5 Gender of the baby the parents don’t want 33 Gender of the baby the mother didn’t want 39 Disappointment because the gender is different to the ultrasound 10 Lack of confidence about their place in the community (e.g after third child) 3 Worry about physical appearance after delivery Negative Stressors 14 Husband isn’t loyal 38 Grandparents don’t accept the baby and send it to an orphanage 30 Husband is rude, drinks, and goes out a lot 13 Lack of care, concern and help from relatives 35 A family member has recently died 45The family observes traditional customs the mother doesn’t want to follow Baby’s Health 6 Stillbirth 15 The baby has a congenital disability 18 Poor health of the baby 32 Baby difficult to nurse 17 Worry about enough milk for the baby 42 Healthy baby 24 The baby is feeding well Mother’s Health 19 The mother has a history of mental illness 8 Obstetric complications (e.g postpartum haemorrhage) 37 Caesarean wound infection 12 Other physical disease during pregnancy and delivery 36 Pain and fatigue after delivery 7 Pain during delivery Protective Factors 16 Sleeplessness due to the baby 28 Regular diet 29 Sufficient relaxation, wakes up late 311 27 Physical activity for avoiding stress 41 Mother doesn’t smoke or drink coffee or alcohol 23 The mother has enough time to take care of the baby Figure 2 Figure 3 312 Figure 4 . in South East Asia and only two community-based surveys in Vietnam, one in Ho Chi Minh City in 1999 and one in Hanoi and Ha Nam in 2009. This study will investigate health worker attitudes. and south of Vietnam. In a sample of 506 women in HCMC, 33% of women were found to have PND, and 19% of participants acknowledged thinking about suicide. In the north of Vietnam, a study of. or depression in Vietnam, and that between 8 and 16.9% of recorded perinatal deaths could be attributed to suicide. There is currently no data on PND from central Vietnam. The purpose of this