Professor Sally Chan, The Nethersole School of Nursing, The Chinese University of Hong Kong, Shatin, Hong Kong. E-mail:

We thank Mr Proctor and Mr Welbourn for their comments on the paper which highlighted cultural influence on the experiences of women with postnatal depression. We agree that perceptions allied to postnatal depression might be affected by other characteristics associated with depressive disorders and obstetric complications during delivery. In our study, all the women who participated were not at their acute phase of depression. The time that they had consulted the psychiatrists ranged from six to 12 months. None of them had been hospitalized for their depression. To ensure homogeneity of the population studied, all of the participants had normal delivery. Women who had obstetric complications were not included in this study.

Proctor and Welbourn commented on the small sample size of the study and recommended using ‘larger sample of peoples experiences, which, when correlated, could provide some insight into the underlying mechanisms of postnatal depression’. When discussing the sample size in qualitative study, Morse and Field (1995) suggested two principles guiding qualitative sampling: appropriateness and adequacy. The appropriateness of the sample is ensured by identification and use of participants who can best inform the research according to the theoretical requirements of the study. Adequacy referred to the stage that data collected are available to develop a full and rich description of the phenomenon. In our study, purposive sampling was used which focused upon women who had been diagnosed as suffering from postnatal depression, and who were willing and able to talk about the experience, thereby increasing the likelihood of acquiring rich data from each participant. The researchers also collect data to a point hat no new data were emerged from the interviews. With reference to Morse and Field's principles, the sample size of our study was considered appropriate and adequate.

The issue of subjectivity in interpreting qualitative data is always a concern in qualitative study. Several measures were employed in our study to maintain objectivity and credibility. Two researchers in the research team were responsible for data analysis. We collaborated closely in cross-analysing in order to achieve consistency and agreement at each step of the data analysis to validate the findings. We also attempted to identify and reflect upon our personal preconceived ideas or expectations regarding the participants’ responses, together with our personal constructs of postnatal depression. We attempted to bracket our experiential knowledge in order to interpret accurately the reality described by the respondents. As suggested by Lincoln and Guba (1985), member checks were performed to enhance the credibility of the findings. Three participants were selected randomly and they were asked to review the description of the findings to confirm that it accurately reflected the essence of their lived experiences with PND. All agreed that the themes generated from the data were accurate reflections of their experiences.

The findings of this study shared some similarities with those found in Western studies (Beck 1992, 1993), in which feelings of helplessness and hopelessness, loss of control, ideas of infanticide and self-destruction were common. We agree on the importance of cultural relativism in the maintenance of cultural uniqueness and that evaluation should be undertaken in light of cultural values, standards and practices. In our study, some findings were strongly influenced by Chinese culture. For example, in Chinese Society, although the family is an important source of social support to individuals, it can also be a burden and source of unhappiness, and is thus a double-edged sword (Kuo & Kavanagh 1994). It is possible that the presence of a family member can induce stress to a new mother, as in this study, where relationships between participants and in-laws were perceived as the major cause of unhappiness. Slote and DeVos (1998) highlights the dominant position of the mother-in-law has its root in Chinese Confucian societies. Although all Chinese societies were male dominant, within the home it was the mother who was the primary force. The primary emotional tie was between mother and son, not husband and wife. The relationship between mother and son is thus firmly entrenched. The daughter-in-law is regarded as a stranger in the house who takes away the son's love from the mother. Slote and DeVos maintain that in families where the attachment of son to mother is strong, daughter-in-law and mother-in-law conflicts are common. These conflicts are not unique to Hong Kong, but are also reported in other Confucian societies.

The majority of participants in our study perceived their husbands as uncaring and controlling. In Hong Kong, traditional gender roles continue to prevail in the family: the man is the breadwinner and the woman the caregiver, even if both spouses are employed. Men are the decision-makers and women the home-makers (Westwood et al. 1995). Many of the marital problems might have already presented before childbirth which surface at the postnatal period. The dominant position of the husband might make it difficult for the wife to ask for help.

In our study, participants had difficulties in doing things against the family's wishes. The obligation of respect and obedience to senior members of the family remains a traditional value strongly adhered to in Hong Kong. The value of filial piety is still strongly held among Hong Kong Chinese. This is influenced by Chinese traditions of emphasis on harmony, interdependence and loyalty, which are different from Western ideals of competitiveness, independence and change (Kuo & Kavanagh 1994).

We agree that experience of women who had depression not associated with childbearing might share similar experience with women who had postnatal depression. Our study was conducted simultaneously with another study on the qualitative experience of women who had depression not associated with childbearing. These women also felt being trapped in the situation. This group exhibited more severe signs and symptoms of depression and had more hospitalization. Divorce and separation were major life events that precipitated their depression. This group of women also did not seek help until their condition became very severe. They also perceived themselves as powerless to control their situations. We found that non-postnatal depression group shared similar presentation of depression as postnatal depression group (Chan et al. 2003).

One of the important findings that needs much attention was both groups of women did not take the initiative to seek medical help for their condition. Access to supporting network or resources outside the family is not a common practice among Chinese women in Hong Kong. In Chinese culture, self-discipline is regarded as the mainstay of social identity and behaviour, and self-esteem results from the knowledge that one is fulfilling one's social role with grace and dignity, and meeting expectations (Kuo & Kavanagh 1994). A woman experiencing depression might perceive failure in meeting their role at their motherhood, marriage, home or work, and consequent loss of face, and shame. Women and their families might not wish to be identified as not coping with their roles and being perceived as inadequate. Some mothers might also worry that they would be labelled as mentally ill, and therefore unfit to be mothers. Women with depression not associated with childbearing might also worried that they were perceived as failure in meeting their role, thus influencing their decision of not seeking help at the early stage.

In another study in Hong Kong by Lee et al. (2000) on identifying psychosocial risk factors for postnatal depression among Hong Kong Chinese women found that postnatal depression was associated with depression during pregnancy, elevated depression score at delivery, and prolonged postnatal ‘blues’. Other correlates of postnatal depression were temporary housing accommodation, financial difficulties, two or more induced abortions, past psychiatric disorders (including depression), and an elevated neuroticism score. Postnatal depression was more likely if the spouse was disappointed with the gender of the newborn. Some risk factors are similar to those found in the West, whereas others (spouse disappointment and history of abortion) may be unique to the Hong Kong Chinese population. We agree that cultural specific assessment and intervention have to be developed for Chinese women with depression.

Proctor and Welbourn commented that more generalized depression screening tools such as the General Health Questionnaire (GHQ) and the Beck Depression Inventory (BDI) developed in non-puerperal context had been found unreliable in detecting postnatal depression. However, Lee et al. (1997, 2001) had compared GHQ, BDI and Edinburgh Postnatal Depression Scale (EPDS) in Hong Kong for screening psychiatric morbidity after childbirth, they found that GHQ, BDI and EPDS were all useful for detecting postnatal depression among recently delivery Chinese women. All instruments have similar satisfactory sensitivity and positive predictive value in detecting postnatal depression. More studies need to be carried out in testing the psychometric properties of these instruments on various cultural groups.

Our paper did not report the details of the physical and psychological treatment that women received in the Postnatal Clinic as it was not the main purpose of this study. We agreed that counselling delivered by trained care workers were effective in helping the women. The Postnatal Clinic in Hong Kong offered counselling service with interpersonal psychotherapy as the main approach. It involved 12–16 weekly face-to-face sessions which was provided by a mental health nurse. The interpersonal therapy approach was used based on the assumption that women suffered from postnatal depression experienced relationship problem with significant others as well as role transition problem. We concur that cognitive behavioural and family focused approaches would be useful to this group of clients as well. Studies have to be conducted to assess the appropriateness of various counselling approaches in different cultural groups.

One of the limitations of our study is that family members were not interviewed. Future studies should consider the perceptions of husbands and family members towards women's experience. Finally we would like to thank the Journal of Clinical Nursing for providing a forum for discussion of cross-cultural issues which are so important in client care.