Received: 13 September 2006
Accepted: 18 December 2006
South Asian women in Britain are less likely to use contraception than women in other ethnic groups. Previous studies have identified a lack of knowledge combined with low levels of English language and/or literacy as barriers to using contraception, but have not examined in detail women’s experiences of accessing services. This qualitative study focused onthe experiences of 19 Muslim women of Pakistani ancestry and the views of six health and community workers. The findings detail considerable institutional barriers to accessing contraceptive services, such as a lack of information and the paternalisticattitudes of some health professionals. The study suggests that, although all the women were motivated to access and use contraception, their ability to make informed choices was often limited. It was only when the women encountered advocates, who might be professionals or from their social networks, that they could begin to take control of their fertility. This study is consistent with earlier research and shows that lack of access to contraceptive services can have high personal and social costs for South Asian women.
Asian women, contraception, service provision
In contrast to white and African-Caribbean women in the same age bands, South Asian women in Britain, that is to say those whose origins lie in the Indian subcontinent, have much lower contraception usage rates (Hennick et al, 1999; Macfarlane and Mugford, 2000; Saxena et al, 2002). The 1998 General Household Survey revealed differences between women of Indian and Pakistani/Bangladeshi ethnic origin. For example, of sexually active 30–49 year olds, 29% of Pakistani/ Bangladeshi women were not using contraception, compared to 15% of Indian women (Macfarlane and Mugford, 2000). These figures do not distinguish between those trying for pregnancy and those with an unmet need for contraception. The 2002 General Household Survey included questions on contraception for the first time since 1998, but data published so far do not give sufficient details of usage by different ethnic groups, particularly in relation to people of Pakistani and Bangladeshi origin. However, Saxena et al (2002) found 60% of Asian women non-users did not intend to become pregnant, and 13% had previously terminated a pregnancy.
There are few studies exploring what influences Asian women’s contraceptive use in the UK. One example is Baraitser (1999) who recruited South Asian women attending a family planning clinic, asking them to identify barriers to contraceptive use. The respondents reported few barriers themselves, but identified problems faced by other South Asian women. They described a division between ‘modern women’ like themselves, who understood and used contraception, and ‘traditional women’ who did not. Recent migrant women were considered likely to be ‘traditional’, whereas UK-born women were likely to be ‘modern’ (Baraitser, 1999, p. 139). Baraitser’s respondents felt the main barriers faced by ‘traditional’ women were a lack of information, embarrassment and language issues.
In another example,Hennick et al (1999) interviewed non-professional and professional South Asian women. In this study many of the non-professional women only acquired detailed knowledge of contraception after marriage or childbirth. Many learnt about contraception from their husbands, and experienced family pressure to have a child shortly after marriage. In contrast, the professional women did not experience the same level of family pressure and often delayed becoming pregnant. While Hindu and Sikh women reported that religion did not influence their contraceptive decisions, this was not the case for Muslim women. The professional Muslim women described their religion as discouraging contraceptive use, while the non-professional Muslim women reported it as prohibited (Hennick et al, 1999). This study outlined issues that may restrict choices for some South Asian women but did not directly address the issue of accessing contraceptive services. This highlights an omission in the literature in that women’s experience of accessing and using contraceptive services has been largely ignored. This paper presents the outcome of a small qualitative study intended to address this topic in relation to women of Pakistani origin in the UK.
The aim of this project was to investigate the attitudes and experiences of South Asian women towards contraceptive service provision. The study was approved by an NHS local ethics committee.
The project team recognised the need for sensitivity, bearing in mind the history of racist assumptions that South Asian women should have fewer children (Katbamna, 2000), and the legacy of coercive population policies in South Asian countries (Lane, 1994). Research could so easily be used to further racist stereotyping, but it is also possible for discrimination to ensue from a lack of systematic inquiry. Our intention was to focus solely on women’s perspectives of contraception rather than make any judgements about whether they would or should have used it.
The interviews used a life history approach. The women were encouraged to start by explaining how they first became aware of contraception, and to situate their attitudes, experiences and decisions within the context of their lives at the time.
Seven South Asian community groups were approached in Coventry, a small city in the West Midlands.Coventry has established South Asian communities, and they make up 11.3% of the city population (National Statistics, 2002). The total number of women in these seven groups was 100. Three groups agreed to participate: an English language class, a sewing circle, and a South Asian mother and toddler group. Two of the groups we approached allowed an explanatory session but decided not to participate; two other groups did not facilitate an explanatory session and did not take part. In the five explanatory sessions, two researchers (PL and RS) attended the groups to explain the project verbally with the aid of interpreters, and distributed the information leaflets which had been translated into Punjabi, Urdu, Bengali and English.
Seventeen female health professionals and community workers who worked in predominantly South Asian areas, andhad professional responsibility for either community support or contraception services within their organisations, were also approached. Eleven declined to participate, giving pressures of time (eight) or feeling that they had nothing to contribute (three) as their main reasons.
Appointments were made with women who indicated interest in participating for the week following the explanatory session. At the beginning of each interview, written consent was obtained (consent forms were available in Punjabi, Urdu, Bengali and English). Nineteen interviews were achieved with South Asian women (aged 18–45 years) of Pakistani ancestry (see Table 1). Four women were interviewed at home, but the majority chose a private room in the community group venue. Sixteen women gave permission to taperecord the interviews, and handwritten notes were made during the other three. All tape recordings were transcribed verbatim, and pseudonyms are used in this paper to preserve confidentiality.
These interviews were based on a topic guide that allowed a sensitive and flexible approach exploring the uniqueness of the women’s accounts, yet retaining a structure for comparison (Rubin and Rubin, 1995). Topics included the acquisition of contraceptive knowledge and experiences of service provision.
Six interviews were undertaken with female health professionals and community workers. These included two South Asian GPs, two white practice nurses and two South Asian community workers. These interviews focused on current service provision. All took place at the informants’ workplaces, but were difficult to arrange. This paper focuses mainly on the themes arising from women’s interviews, but draws on the professional and community workers’ interviews where this will help to contextualise the issues.
There has been an ongoing debate about the necessity of matching interviewer and informant’s ethnicity. For example, Rana et al (1998) suggest that South Asian interviewers are more likely to achieve rapport with Asian informants, while Rai (quoted in Hennick et al, 1999) argues that if interviewers are from a non- South Asian background, informants may be more reassured about confidentiality in sensitive areas. In this project both South Asian and white interviewers were made available, but only one person stated a preference, which was for a non-Muslim interviewer due to concerns about confidentiality.
Interpretation is not just a technical issue, but can impact directly on the research findings (Temple, 1997). Five of the interviews with the women were all or mainly in English, and five in Punjabi. In the other nine interviews a female Urdu interpreter was employed to assist the researchers. The six interviews with health professionals were conducted in English. As a quality control measure, the Urdu interview taperecordings were transcribed by another Urdu speaker, who was asked to highlight where there could be differences in interpretation. While back-translation may not reveal all interpretation issues (Bradby, 2002), it showed that while on the whole both interpreters agreed on the interpretation, the Urdu interpreter had often taken a pro-contraception stance rather than a more neutral stance intended by the researchers. This has been taken into account during the analysis by, for example, considering if phrasing a question differently may have had an impact on the response.
A modified grounded theory approach (Strauss, 1987) was used whereby the transcripts and interview notes were coded and analysed for emerging theory, but theoretical sampling was limited in order to act in accordance with the ethical approval. The team traced the history of each woman’s contraceptive usage, including the events each woman felt to be significant, drawing a ‘time-line’ for each woman. This was used to clarify the woman’s story and how the data subsequently coded by theme were positioned within each story. The data were coded for themes, initially those relating to the interview schedule. However, as familiarity with the data grew, other themes emerged. The data were reassessed by re-reading the transcripts and field notes, and for each theme the transcripts were searched both manually and using keyword searches for relevant sections of the interviews. Separate documents were produced for each theme containing all the identified sections of the data to allow systematic comparison. The health professional interviews were thematically analysed using a similar process, and the data compared to the women’s interviews (Strauss, 1987).
Nine of the women began their married lives with little or no knowledge of contraception, and only began to use it following the birth of a child or children. Four women reported some education at school (two in the UK, two in Pakistan), and six had been given some information by female relatives, but none of them felt they had been told enough.
Although an extreme case, Soraya’s story highlights issues for women lacking contraception information. Soraya was 16 years when she married and migrated from Pakistan to the UK. She quickly became pregnant, and this child was rapidly followed by two more. It was only after the birth of her third child that Soraya recalls anyone mentioning contraception:
‘I was really frightened ... to keep having the babies so quick ... because it is 11 months different between them three ... I didn’t know if I can use anything, or my husband can use anything ... I didn’t know anything about that ... the midwife, after the baby, when she came to check me ... and she told me ...’ (Soraya, aged 26 years)
Following her third child, Soraya used contraception consistently for six years, but had now decided to try for another baby. She said none of the health professionals she had seen had provided her with any contraceptive information. Nasreen and Farida also had more than one pregnancy in the UK before hearing about contraception. When asked if anyone had talked about contraception in the hospital, Farida said she ‘didn’t know’ as she did not understand what the nurses were saying.
Most women discuss contraception with other women in their social networks and combine this source of information with leaflets or advice from health professionals (Meadows, 2001). Our informants felt they had limited knowledge, and some reported being solely reliant on friends or family for contraceptive advice, due to the lack of interaction with health professionals. General practitioners (GPs) were identified as their initial access to contraception, with women finding out about family planning clinics later. Even then, some women did not know that they could make their own appointments at a clinic without being referred by their GP.
When asked what would make contraceptive services easier to access, most women said that they needed more information, particularly within community settings. Sakina suggested that an increase in ‘classes ... in women’s health’ would help women be more confident. This view was endorsed by one of the community workers:
‘I think outreach is really important ... Reaching out to where they sort of meet ... because I am sure ... I mean when, if you go and speak to them ... they do want to access it, it is sometimes they do not know enough, but when services come out and meet with them ... that increases more people from the ethnic community to access it.’ (South Asian community development worker)
Most of the women interviewed appreciated the outreach health education in community groups, and all the health professionals and community workers emphasised its importance. However, many of the women had not been attending community groups at a point when they first needed contraceptive advice, so this service was not always reaching women at the appropriate time.
Thirteen of the women interviewed had experienced problems communicating with their health professionals. In our study, although language barriers were mentioned by women, they described other issues as more problematic. Analysis of the interviews revealed three particular issues for these women: language, gender and paternalism.
Seven women described gaining confidence in talking to health professionals through repeated contact. Jakeen’s account was typical:
‘In hospital because of operation, I stay in one week in hospital, that helped very much. I try my best to talk in English, I got confidence, little bit ... after my daughter born, after this I got confidence to go myself own ... I am not going with my husband, I am not going with my cousins ... In pregnancy you know ... we have to go again and again to surgery ... so that’s why I learn much ...’ (Jakeen, aged 26 years)
Previously, they had mainly relied on friends or family to interpret in health consultations. About half of the women were happy with this situation; for example, Rahima mentioned that she discussed everything with her sister-in-law anyway, so she was happy with her being present. However, others were uncomfortable taking someone with them. Nasreen found having her husband present was embarrassing at times. Other womenwere concerned about the inconvenience caused to relatives, and would try to manage alone, even when they were worried about communicating.
Previous research has identified problems with using family as interpreters (see, for example, Bhakta et al, 2000). Our study confirmed that for somewomen, using family interpreters was problematic. For example, one of the participant GPs who spoke Urdu had encountered a mother-in-law describing her recently married daughter-in-law as having problems becoming pregnant. When the GP discussed the issue in private, the younger woman did not want to become pregnant yet, and asked for contraceptive advice. The privacy granted through the GP’s language skills was the key to enabling the younger woman to access her preferred contraceptive method.
One of the practice nurses was concerned about how interpreters could modify the emphasis of contraceptive information. Although this can always be an issue in interpretation, including in the interviews themselves, at least professional interpreters can receive training about such issues. However, despite growing recognition that relying on friends or family is often inappropriate, none of the women had ever been offered an NHS interpreter. As one of the GP participants pointed out, although the use of interpreterswas growing, they were rarely used for South Asian women:
‘I have been here for 10 years now, and it is only in the last few years the use of interpreters and translators has really shot up ... I mean I had two in yesterday ... and yet Asian women have been coming here for years, and don’t speak English, we have never had translators for them, and still we don’t.’ (GP participant)
The lack of NHS interpreters can put extra pressure on bilingual health professionals. A GP who spoke Punjabi as a second language described how she did not have the vocabulary to explain everything. Conducting consultations in Punjabi was an additional stress for her, and so she limited the sessions where she was likely to need to do this.
Like many other women, most of our informants preferred to consult female health professionals about contraceptive services (Lowe, 2005), particularly for intimate examinations, but not all of them knew they could request this. Soraya missed appointments when an examination was likely if she thought the doctor would be male. In other cases, the women were taken by surprise by the sudden involvement of a male doctor. For example, Shaheeda had not known that her intrauterine device (IUD) fitting would be with the male doctor, as she was initially seeing the nurse. She stated:
‘I had an appointment, the nurse got the coil ready and everything ... I went in, and she said I am going to call the doctor to put the coil in for you, and I said to them, no way am I going to have a doctor ... I’ll have 10 kids, but I will never have that from the doctor.’ (Shaheeda, aged 32 years)
In this case, while there was a delay, her request for a female doctor was met. Khalida did not feel able to protest, despite what she perceived as the ‘sudden’ involvement of a male doctor in an intimate procedure, and has remained unhappy about this since.
Despite consulting doctors who spoke their language, the women often found it difficult to ask questions about contraception, particularly if the doctor was older and male. Shaheeda stated she was ‘like a daughter’ to her GP. Cultural issues of respect meant that the women often felt unable to insist that their voices were heard. Two cases illustrate the devastating consequences for women. Shaheeda described having severe complications during her fourth pregnancy and that her consultant advised her not to have any more. Despite this her GP refused her request for sterilisation. Having had problems with the combined pill and an IUD, Shaheeda repeatedly used emergency hormonal contraception (she estimated at least ten times), and had three terminations in quick succession. She commented:
‘I said to the doctor, third time I got pregnant, third time I had abortion ... why are you not letting me do what I want? ... but he says no ... he is a doctor, and mostly he is right, maybe it will affect me ... maybe it is dangerous for me.’ (Shaheeda, aged 32 years)
Shaheeda wondered if her GP was hinting that it would adversely affect her health. However, the constant concern about becoming pregnant was also affecting her. She described her despair:
‘I know I would never had that child, I wasn’t regretting that I had the abortion ... I was thinking ... I was so stupid to get pregnant ... because I couldn’t take any contraception ... it affected me badly ... I used to pull my hair ... and cry and cry ...’ (Shaheeda, aged 32 years)
Shaheeda attributed her GP changing his mind about the sterilisation to the intervention of health professionals at the abortion clinic. She was very happy that the procedure had finally taken place.
Farah, aged 35 years, became pregnant for the sixth time after her request for sterilisation was denied by her GP. He also discouraged a termination on the grounds of their shared religion. During routine hospital antenatal care, Farah mentioned her wish for sterilisation and the consultant placed her on his waiting list. She was told that if she had requested this during her fifth pregnancy, they could have put her on the waiting list earlier. Farah blamed herself for not being sterilised earlier:
‘I didn’t know, at that time, I didn’t know you see ... somebody told me you had to tell your GP, I was all like confused ...’
Both Shaheeda and Farah reported that their GPs told them that consultants would not accept a referral because they were under 30 years, but that the hospital denied this policy.
Other women reported that health professionals made decisions about contraception for them, and an IUD was particularly recommended.One of the South Asian community workers interviewed felt that some GPs weremore concerned about likely compliance than women’s preferences. Previous research has revealed that health professionals tend to advocate methods of contraception that do not rely on self-administration to women they feel are ‘irresponsible’ (Todd, 1984; Foster, 1995; Hawkes, 1995).
Despite difficult experiences when trying to access contraception, the women were, in general, uncritical of the health professionals’ behaviour. Zaida’s account is typical of how women talked about their health professionals. Zaida had gone to her GP to request emergency hormonal contraception:
Zaida: ‘He first shouted at me ... you have no coil, no pills you no stop children ... they [her children] making a lot of noise, they are jumping here, there ... the doctor said ... you can’t look after them two ... and how are you going to have another one now ...?’
Interviewer: ‘Did he upset you, the doctor?’
Zaida: ‘... no, no, not upset ... he is very nice.’ (Zaida, aged 34 years).
The term advocate here is used to describe a person who offers support to others, rather than speaking for them in any particular circumstances. The women we interviewed had gained confidence in being able to access contraceptive services. Encouragement from another person, usually another woman rather than a health professional orcommunity worker, seemed to make the difference. It usually took time, involving several conversations and hearing stories, often at community groups. Positive experiences of services were crucial to a recommendation. In contrast to Hennick et al (1999), none of the women mentioned their husbands as sources of encouragement in gaining their chosen contraception. The women interviewed had begun to encourage others. Farah commented:
‘I see ladies, they don’t go to doctor, if they have private problem ... they won’t want to talk to the doctor ... they say we can’t ... I have told them to go to the clinic ... you know the women’s clinic at the hospital, I tell them to go there ... and then they say to me, ‘‘oh we feel relaxed now’’.’ (Farah, aged 35 years).
Lay advocates are thus an important source of information and support for women, but they may not be encountered at appropriate times.
This study took a life history approach to understanding the experiences of women of Pakistani origin accessing contraception in the UK. This approach has highlighted how women’s needs change over time as their situation changes, and how this, coupled with a lack of awareness of their entitlements within the UK NHS, can make it very difficult for them to achieve their choice of contraception.
All of the women had become confident contraceptive users, yet over half of them had experienced an unintended pregnancy. Culturally sensitive information and encountering an advocate at a time that was personally pertinent were important factors in achieving confidence, a finding consistent with previous research (Meadows, 2001). Consequently, women are more likely to learn about contraception when they are considering using it rather than in general informationgiving sessions. At a community education level, learning how to access contraceptive information may be more important than being given precise details on methods themselves.
The women encountered a number of difficulties in accessing services. Language barriers remained important, but, like Baraitser (1999), we argue that other issues ought to be considered. When accessing services, women needed to be able to trust whoever was interpreting for them, particularly with regard to sensitive areas. Interpreter choice is a complex issue and may vary depending on the context (Baraitser, 1999). Offering professional translation services is important, but the final decision as to who interprets should be left to the women themselves. Being unaware that they could request female health professionals also acted as a deterrent. Women sometimes had to choose between a male health professional who spoke their language and a femalewho did not, and, in addition, thewomen were not always aware of different professional roles. These issues are further complicated by the embodied nature of trust in contraceptive consultations: female health professionals are generally considered to be more able to give advice (Lowe, 2005).
Few of the women felt they had been given sufficient information by health professionals. Insufficient information is likely to lead to discontinued use of contraceptives (Walsh, 1997). A strong recommendation from a health professional could deter women from asking about alternative methods of contraception. The paternalistic attitudes of some health professionals caused difficulties for some women. White British women often encounter similar problems, but their better knowledge of entitlement allows them to overcome them, usually by switching to an alternative provider (Lowe, 2005).
All of the women had experienced times when they were unable to exercise their desired control over their fertility. Although a small study, the similarities of these findingswith previous studies (Baraitser, 1999; Saxena et al, 2002) suggest these are not isolated cases. The two UK-born women encountered very similar problems to the women who had migrated. Through taking a life history approach, this study elicited the experience of women before they became competent contraception users. The women’s descriptions of their early difficulties may well be coloured by subsequent experience. Although this needs to be taken into account, the stories told by the women provide a complementary account to that of other qualitative studies. For example, Baraitser’s (1999) accounts of the difficulties facing migrantwomen were second-hand. The experiences recounted by the women were often some years in the past. It is possible that given the same circumstances now, the women may have had a different experience. Unfortunately the interviews with the health professionals would indicate that this may not be the case, although this does not preclude the possibility of excellent services in certain localities.
Although Hennick’s study (1999) found that religion influenced women’s access to contraception, in this study, none of the women interviewed felt that contraception was prohibited or discouraged by their religion. This may be because of self-selection in the recruitment process. Issues of izzat (family honour and shame) were not reported by the women as impacting on their contraceptive use. Although many women find it difficult to voice their concerns, particularly to male doctors (Lowe, 2005), as izzat impacts on most areas of Muslim women’s behaviour (Afshar 1994), this is likely to have made it more difficult than for other women to make their requests.
This project adds to previous studies which have identified considerable levels of unmet need for contraception among South Asian migrant women in the UK. Further research is needed to clarify the extent of this unmet need and whether it is found in other minority communities. While not all unintended pregnancies are unwanted, there can be a high personal and social cost incurred when appropriate contraceptive services are not accessible.
Our thanks go to the women who shared their lives with us, the community workers who supported this research, and to Coventry PCT for their support. We would also like to thank the reviewers for their helpful comments. This project was funded by Warwick West Midlands Primary Care Research.