Introduction: background and significance
1The levonorgestrel emergency contraceptive pill (referred to from here on as EC, meaning the dedicated product Postinor or Norlevo, not the regular contraceptive pill or Yuzpe method) is one of the latest additions to the global contraceptive mix, but its use is controversial. The anti-abortion lobby claims that its mechanism of action is abortive, while clinical evidence suggests that it simply delays ovulation. This, together with the fact that it is only used post-coitus has sometimes led to a politicization of debate on EC use. Its position within a traditional family planning programme – whether it should be accepted as part of the regular contraceptive mix, included as a method of last resort or rejected entirely from regular contraceptive provision – has not been well studied. This is particularly true in sub-Saharan Africa where regular contraceptive use is often limited.
2To be effective, EC must be used within 72 hours of unprotected or poorly protected sex (Free et al., 2002), although a new product (ulipristal acetate, not yet available in our study countries) can be used up to five days afterwards (Glasier et al., 2010). Yet the people at risk and the providers to whom they turn are not always aware of this time constraint (Teixiera et al., 2012). Even once people decide to seek and use EC, there are many barriers associated with EC provision. Health care providers are often influenced by prevailing social norms on acceptability of EC, for example the extent to which it is seen as abortive (views on this may be compounded by lack of knowledge), and who should have access to it. This study seeks to understand providers’ knowledge of EC, their attitudes towards its use and how they respond to users seeking care or help.
3It has been shown that inadequate provider knowledge is an important barrier to EC provision and use in a range of settings (Nigeria: Ebuehi et al., 2006; Fayemi et al., 2010; Northern India: Tripahi et al., 2003; Turkey: Sevil et al., 2006; Aksu et al., 2010; Laos: Sychareun et al., 2010; and Europe: Uzuner et al., 2005; Szucs et al., 2010). Studies from the United States indicate that even “knowledgeable” providers do not always have accurate knowledge of how and when EC should be taken or of its mechanisms of action (Sherman et al., 2001; Wallace et al., 2004; Lawrence et al., 2010; Farris et al., 2010). Even where providers do have adequate knowledge, regular supplies and even favourable attitudes, they may still under-prescribe (Sherman et al., 2001; Wallace et al., 2004). Studies have found that providers seldom educate their clients about EC and do not proactively assess their need for EC in routine health and family planning visits, either in Northern countries (Delbanco et al., 1998; Karasz et al, 2004) or in Africa (Ebuehi et al., 2006), despite evidence from the United States that if providers initiate EC discussion women are more aware of when to use it (Harvey et al., 1999).
4Provider attitudes are influenced by a range of professional and moral factors which have been documented in relation to a wide range of sexual and reproductive health services, including emergency contraception. Access to services is frequently reduced as a consequence, especially for young and unmarried people in Africa (Mantell et al., 2000; Mantell et al., 2001; Adeokun et al., 2002; Mngadi et al., 2008) and in other Southern countries (Hobcraft and Baker, 2006; Fairhurst et al., 2004). Providers’ concerns include: reluctance to actively promote EC for fear that it will encourage unprotected sex and therefore increase risk of HIV and sexually transmitted infections (STI) (Ziebland, 1999; PPAG, 2001; Karasz et al., 2004); concern for health side-effects of frequent use of high-dose pills (PPAG, 2001); and concerns about the potential for regular EC use to replace regular family planning, though this has been refuted (Marston et al., 2005).
5Reflecting the socio-cultural context in which providers work, normative moral concerns emerge as frequent barriers to timely provision of EC (Davidson et al., 2010). The most widely aired of these is the possible association of EC with abortive-type mechanisms of action – a view most notably advanced by the US Food and Drug Administration under George Bush (The Lancet Editorial, 2005; Ackerman, 2006). There is some evidence that providers around the world may share this view. Glazier et al. (1996) found that providers in Malaysia were unwilling to provide EC because they thought it might be abortive. A study of member associations of the International Planned Parenthood Federation (IPPF), leaders in EC provision, found that more than half of providers in Southern countries did not provide EC, partly because of fears over abortive mechanisms (Senanayake, 1996). According to several American studies, providers fear that promotion of EC might suggest to young people that unprotected sex was acceptable (Ziebland, 1999; Karasz et al., 2004). Rutgers and Verkuyl (1998) found that EC providers in Zimbabwe were particularly disapproving with respect to adolescents. A study in the United Kingdom found that pharmacists were preferred by users since they offered a less judgemental service (Bissell and Anderson, 2003).
6A major drawback of the extant literature on providers’ knowledge, attitudes and provision of EC is the dearth of studies from Southern countries. We know very little of how providers in Africa interpret and understand emergency contraception, whether this affects their service provision behaviour, or whether and how they judge different types of clients presenting for EC. EC has great potential to contribute to reducing abortions and maternal mortality in sub-Saharan Africa, particularly West Africa, where family planning use is very low and unwanted pregnancies, subsequent abortions and maternal mortality rates are high (Ghana DHS, 2009; INSD and ORC Macro, 2004). However, despite the importance of this question, we have found only one study on the views and attitudes of providers towards EC in sub-Saharan Africa (Fayemi et al., 2010); along with a few partial exceptions, now quite old (Gichangi et al., 1999; Muia et al., 1999; Muia et al., 2000; Steiner et al., 2000).
7The study reported in this paper is the first to be based on in depth interviews seeking to ascertain the knowledge, attitudes and practice of reproductive health providers regarding emergency contraception in two West African countries, Ghana and Burkina Faso.
8After outlining the methods, this paper describes the context of family planning programmes, EC availability and changing socio-cultural contexts for contraceptive use in the two countries. The findings section begins with a brief description of the availability and knowledge of EC among providers. The paper then analyses the major dimensions that emerged from the interviews and were used to construct a typology of providers’ attitudes and responses to EC provision.
I – Methods: semi-structured interviews with providers in Burkina Faso and Ghana
9This study is part of a broader project named Emergency Contraception in Africa (ECAF) which took place in four countries: Ghana, Burkina Faso, Senegal and Morocco. Interviews with providers were conducted in all countries, but only Ghana and Burkina Faso were selected for discussion in this paper. Morocco has a completely different social and religious context and was not deemed comparable to the other three West African countries; in addition a dedicated EC product (as opposed to estrogen-progestogen contraceptive pills used for the same purpose) was not legally available in the country at the time of the survey, so there was little discussion of EC and its use in the provider interviews. In Senegal the interviews were conducted and translated too late to be included in this analysis. Ghana and Burkina Faso, on the other hand, are neighbours in West Africa but provide an interesting contrast given their different colonial, political and social development. In both countries a levonorgestrel-based EC product is available – since 2000 in Ghana and since 2003 in Burkina Faso. The study was conducted in the capital cities only, because this is where EC was available without a prescription.
1 – Sampling design
10The purposive sample included EC providers from public and private sector facilities (including pharmacies) who were selected from government lists based on their proximity to the centre of Accra in Ghana and Ouagadougou in Burkina Faso. Their characteristics are shown in Table 1. Most respondents (in both countries) were women; in Burkina Faso, several respondents were doctors, indicating the greater medicalization of contraception in that country compared to Ghana, where doctors do not usually provide contraception. We may not be sure to have reached saturation (due to time/budget limitations), but the clear groupings that emerged in our typology indicate a broad consensus in the findings consistent with saturation.
Characteristics and number of EC providers who took part in the ECAF survey in Ghana and Burkina Faso
Characteristics and number of EC providers who took part in the ECAF survey in Ghana and Burkina Faso
11In Ghana the interviews were conducted in English and transcribed; in Burkina Faso they were conducted in a mix of French and local languages, transcribed into French, then translated into English. All interviews were analysed using the qualitative software package Atlas-Ti v.6. They were coded and analysed for emerging themes which were analysed and interrogated.
12Ethical approval was obtained in Ghana (Ghana Health Service Ethical Committee), in Burkina Faso (Comité d’éthique pour la recherche en santé), and from the ethics board of the London School of Hygiene & Tropical Medicine.
2 – Qualitative Interviews
13Thirty-one semi-structured interviews were conducted with providers of reproductive health services in Ghana and Burkina Faso in 2007-2008. Interviews used a detailed topic guide (the same in the two countries) to cover providers’ knowledge and views of different forms of emergency contraception, what types of EC clients they saw and how they responded to them, as well as more general questions on contraceptive service provision. The guide was flexible and did not require that a rigid order be followed, or specific questions asked. Rather, the interview was structured around the respondents and the issues they spontaneously raised in the interview. Interviews were conducted by a single member of the research team in each country (both female). The guides were pre-tested in both Ghana and Burkina Faso.
14In Ghana, the interviewer was a member of the Ghana Health Research Unit, while in Burkina Faso she was a member of the University of Ouagadougou. In both countries we ensured that in their dress, introduction and conduct they did not present themselves as in any way “superior” to the interviewees, and at the beginning of each interview the respondents were told that we were interested to hear their experiences, opinions and views, and that there were no “right” or “wrong”’ answers. Interviews were conducted in the language chosen by the interviewee. Both interviewers had been trained in qualitative techniques, already had experience of conducting qualitative interviews, and were involved in discussion about the research itself, its purpose and mission. We are therefore confident that any potential barriers caused by social hierarchy were minimized and that the quality of the interviews was high.
3 – Typology development
15Typologies are usually multi-dimensional and involve classification into mutually exclusive categories (Ritchie et al., 2003; Hammersley and Atkinson, 1995). Our typology of providers included two major dimensions: “acceptance” of EC by providers as a legitimate product, and the “provision-response” of providers to clients presenting in need of emergency contraception. The acceptance dimension encompassed providers’ views on three issues informed by the literature: the mechanism of action (is it seen as abortive?), impact of EC on regular contraceptive use, and how widely EC should be made accessible. From these three sub-categories, an overall acceptance level could be ascertained (negative; concerned; cautious; positive). The provision-response dimension considered whether providers or clients initiated the EC provision, and the manner in which providers responded to different types of clients (not at all; reluctantly; with restrictions; immediately). Before moving to the context and findings, note that interviews with potential users of EC in both Accra and Ouagadougou (100 women and 54 men) were analysed in a separate part of the ECAF study and are written up in a separate paper (Teixiera et al., 2012). Where appropriate, we have drawn on this separate analysis in the discussion section of the current paper to provide a contrast to the findings among providers examined here.
II – Country contexts
1 – Ghana: a pioneer in sub-Saharan Africa
16In West Africa, Ghana was the first country to engage in family planning activities. Between 1975 and 1979, the total fertility rate (TFR) was 6.5 children per woman (Shah and Singh, 1985), falling to 5.5 in 1993 and then to 4.0 in 2008 for the country as a whole, and 2.5 in the capital (Ghana Statistical Service, 2009). Ghana has reached its target of 4 children per woman two years ahead of its target date of 2010. It reached its TFR target despite slightly slipping in its contraceptive prevalence target of 28%, achieving 24% by 2008. To achieve its TFR goal of 3 children per woman by 2020, it is anticipated to need a prevalence of modern contraception of 50% by that date.
17The very first family planning activities were begun by the Christian Churches Council and then the Planned Parenthood Association of Ghana in the late 1960s. In 1969, a population policy was promulgated, highlighting the specific risks facing women, children and young people (high fertility, risky abortion, high rates of morbidity and mortality) and announcing national access to contraception to address these risks. In 1970, the first national family planning programme set ambitious targets to reduce the population growth rate, but progress stalled in the 1980s and 1990s through lack of political will, national economic decline and socio-political turmoil (Locoh and Makdessi, 1996; Caldwell and Sai, 2007).
18In the 1990s, the population policy was revised in line with the International Conference on Population and Development in Cairo in 1994, recognizing that the birth rate was still too high, particularly among adolescents, and that the spread of HIV required attention. The policy objectives were enshrined in the Constitution (Center for Reproductive Rights, 2003). Abortion has been legal in Ghana since 1985 in cases where the life or the physical or mental health of the mother are at risk, in cases of rape or fetal malformation, but this law is rarely applied because both the population and health personnel are largely unaware of its existence. The social marketing programme supported by UNFPA has enabled pharmacists and sellers of pharmaceutical products to sell condoms, spermicides and oral contraceptives (Raylynn, 1994) since 1986. The introduction of combined injectable contraceptives and implants (in 1996), and female condoms (2000) has widened the range of products on offer. In addition, oral contraceptives have been licensed for use as EC since 1996 and the dedicated EC pill itself (Postinor) has been licensed since 2000. In practice EC is easier to obtain in the capital Accra (the focus of this study) where it is available in clinics and pharmacies; a prescription is not needed although some providers will not dispense EC without one.
2 – Burkina Faso: late demographic intervention
19In Burkina Faso, fertility has declined much more slowly, from 6.9 children per woman in 1993 to 6.2 in 2003 (Macro International Inc., 2000; INSD and ORC Macro, 2004) and, as in Ghana, it is much lower (3.1) in the capital Ouagadougou. Currently less than 10% of women in Burkina Faso use modern contraception.
20In 1978, Burkina Faso signed the Alma Ata declaration, whose mother-and-child health component included a section on family planning through birth spacing promoted by midwives. A 1920 law forbidding anti-conception propaganda and abortion was still in force at the time and was included in the Public Health code of 1970, although the code did permit therapeutic abortion to save the life of the mother, and the penal code provided for a theoretical right to abortion in certain prescribed cases.
21The first major action in population policy was the creation in 1983 of the Conseil national de population (CONAPO) to develop a national population policy which was duly instituted by decree in 1986. Its principal aims were to encourage birth spacing, to avoid unwanted pregnancies and to enable couples to choose their number of children (Ministère de la santé, 1986). Partial abrogation of the law of 1920 legalized access to contraceptives within public health institutions but required laboratory tests which were expensive and therefore a disincentive (Ministère de la santé de l’action sociale et de la famille, 1986). This condition was removed in 1988.
22In 1991, the “Politique nationale de la population” set itself the ambitious objectives of lowering fertility by 10% every five years from 2005, raising the contraceptive prevalence rate and lowering the infant mortality rate. Family planning services include contraception and the prevention of STIs and HIV. In the “Code de santé publique” (article 86) it was spelled out that all techniques and methods in family planning, including sterilization but excluding induced abortion, were legal in public and in private health institutions. In terms of emergency contraception, the dedicated product (Norlevo) was licensed in 2003, though oral contraceptives were also used for EC before this. As in Ghana, EC is available through clinics and pharmacies and can be obtained without a prescription, although some facilities require one.
3 – Social contexts and use of family planning in urban Africa
23The social context of sexual behaviour and family planning use is changing in urban Africa. Legislative changes with regard to marriage, the increasing time spent in education, women’s inclusion in the labour market, the context of economic crisis in which male positions become especially insecure (Adjamagbo and Antoine, 2002), have all contributed to changing the conditions of sexual debut and first union and, more widely, of relations between the sexes. In sub-Saharan Africa, age at marriage is increasing, and sexual debut is no longer closely tied to marriage for women. Although virginity at marriage, especially for women, is still prized, it is decreasing in importance and prenuptial sexuality is tending to become more commonplace in urban settings. In both countries, family planning services are officially accessible to all women, although there are still major barriers to their use by unmarried women, the youngest especially (Stanback and Twum-Baah, 2001).
24The contraceptive practice of Ghanaian and Burkinabe women displays some common features: natural methods (23% of methods used) are used as frequently as modern methods (the pill represents 22% of methods used in Ghana and 16% in Burkina Faso; injections 21% and 18% respectively; condoms 12% and 15%). Contraceptive prevalence rates are almost twice as high in Ghana as in Burkina Faso (25.2% versus 13.8% of women use some kind of contraception). In both countries, the prevalence of use of modern methods rises strongly with women’s educational level, their wealth (which may be related to education and access) and residence in an urban area where services are more easily accessible (INSD and ORC Macro, 2004; Ghana Statistical Service, 2009). Condoms represent 80% of all methods used by unmarried women in Burkina Faso and 41% in Ghana (Blanc, Tsui et al., 2010).
25Contraception is thus becoming more widely used and socially accepted in urban settings, especially for sexually active unmarried women (Cleland and Ali, 2006). The role of EC in the contraceptive mix is currently unknown. The ECAF study provides some of the only data on EC use and shows it to be extremely low (Teixiera et al., 2012), corroborating the few other studies that exist (Opoku, 2010).
III – Study findings: acceptance and provision of EC
26The respondents’ characteristics are summarized in Table 3, with a single identifier indicating the country concerned (B = Burkina Faso, G = Ghana) and a number.
1 – Provider acceptance of emergency contraception
Availability and knowledge of EC among providers
27The official availability of emergency contraception is governed by the drugs licensing bodies of both countries where it is legally available. At the time of interview, about half of all providers in Ghana and Burkina Faso said they had current stocks of the dedicated EC pill at their facility. Available data suggest that private providers have more flexibility to go directly to wholesalers or drug manufacturers for supplies, although this didn’t prevent stock-outs entirely.
28Providers in both countries were asked what different forms of EC they knew (oral contraceptives, IUDs, dedicated EC pill or any other). In general, they displayed widespread knowledge: virtually all providers in Ghana and Burkina Faso had heard of EC and most (12 in each country) had given it in some form, primarily the EC pill or oral contraceptives. Accurate knowledge of at least one effective method of EC (i.e. the dedicated EC pill, oral contraceptives or IUDs) was somewhat less frequent than having given EC. Frequency of provision was highest in Ghana: about a quarter of respondents gave it on a daily basis; another quarter gave it regularly through the month; three gave it just occasionally through the year; while three had never given it. In Burkina Faso, few had given it regularly; most had given it occasionally through the year while four had never given it.
Perceptions of emergency contraception and abortion
29In Ghana only one provider thought EC “could be” abortive, but was still in favour of its use and had only refused it to clients if they thought they were already pregnant. None of the other Ghanaian providers thought EC was abortive, even though five mentioned that it worked by “preventing implantation” rather than by preventing egg release/fertilization. One provider indicated that a training workshop had changed her view:
Initially I… I wasn’t too happy about it. I thought it was a form of a termination of pregnancy […]But then I […] attended a workshop on it and I realized that it wasn’t as I thought.
31In Burkina Faso, of the few respondents who did explicitly discuss this question, one (a Catholic) thought it was abortive but she interpreted abortion as preventing the sperm from meeting the egg. One other provider who had read several inconclusive studies was concerned it might be abortive, but said
I had to prescribe it because it’s better to prevent a pregnancy than to see people go for an illegal abortion in doubtful places.
33Many other respondents contrasted EC with “illegal abortions”, suggesting they did not see EC itself as abortive, and all respondents in Burkina Faso were in favour of EC in general, though three of these (including the one who said it was abortive) were reluctant to provide it.
34Eventually three positions emerged: those providers who thought EC was not abortive, those who thought it was, and those who held a contradictory position. In general EC was not considered abortive – or if it was held to be of ambiguous action this was not seen as problematic. For the typology, two responses were thus defined: concern; no concern (which encompassed all who did not express active concern).
Perceptions of emergency contraception and regular family planning use
35In Ghana the general view among providers seemed to be that making EC more widely available would not have a negative impact on family planning (FP) use. Indeed several mentioned EC use as an entry point for promotion of regular FP use, thus seeing it as an opportunity:
We had a client… she thought her husband would go in for vasectomy which he wasn’t willing to do, so she said she would use EC whenever she thinks she is not safe and she should have sex around that time. So we talked to her and later she came in for an IUD.
37Only three providers expressed concerns that EC might in some instances displace regular FP, while a further two recognized it might be an issue but were confident that by counselling their clients they could prevent them from using EC as a regular method. While most providers did not want to see EC being regularly used, a few did recommend EC as a short term method in certain specific circumstances, as illustrated by this provider in Ghana:
If like somebody’s partner [lives away] and he comes in once in a while […] with that the client can be on emergency contraception because after the partner is gone back […] if the wife is not going to have sex, there is no need to be on a contraceptive method…
39One pharmacy assistant even reported using it herself on a monthly basis because “I don’t like the regular pill, maybe I can forget” (G8) although she also said that people shouldn’t do this because of potential health effects.
40Providers in Burkina Faso showed most caution with respect to displacement of regular FP, with eleven providers mentioning some aspect of this issue, though their main concern seemed to be that EC would displace condom use and therefore lead to an increased HIV risk. The following were typical:
[EC] is not a good thing to extend because it will lead people to abandon methods that consist of protection not only against HIV but also from unwanted pregnancies […] by taking Norlevo for example, you don’t avoid the risk of getting AIDS.
Young people won’t use condoms. They will go straight away for it [EC] saying they have had sex.
43Like in Ghana, several providers talked of encouraging their EC users to adopt regular FP methods, thus using this as an “entry point” for family planning, though almost all said they would prefer to promote FP rather than EC.
44For this issue, then, three responses emerged: those who felt EC had no effect on regular FP; those who saw it as an entry point for FP to potentially increase FP use; and those who felt it would displace FP. For the typology these were reduced to two: concern; no concern (which combined “no effect” and “entry point”).
Accessibility and other issues
45Concerns about the health consequences of frequent EC use seemed widespread. For this reason, most providers in both countries preferred to see EC provided through health facilities and pharmacies with trained staff to ensure that correct information on both EC and FP was given. Although some providers wanted restrictions on provision, most were happy for EC to be a non-prescription drug.
46In Ghana, 12 of the respondents thought EC should be available to anyone, although three of these wanted it prescription only. Two wanted it only for over 18s: one because of health concerns (of hormone doses), one because it might encourage promiscuity. In Burkina Faso acceptability was also high, with 12 respondents who thought EC was good for a wide range of groups. Among them, three said they would prefer it for married people but would not refuse anyone. In terms of where it should be provided, in Ghana nine respondents wanted EC to be only provided through proper health facilities with trained staff due to concerns about proper EC use. Only two wanted to see it provided anywhere (e.g. at the hairdresser’s) as is the case for condoms, making them more widely acceptable and accessible, especially for young people:
… the young ones will not come in [to the clinic] […] i if there is more…to the pharmacy shops […] I think it will be better, […] and even like the CBDs (community-based distributors) and then the hairdressers and, ehm, dressmakers. Because we have some that we supply them with condoms and their girls patronize it.
48Providers in Burkina Faso seemed much more cautious. Many respondents expressed their concern that “popularizing” EC would increase its use, and that this would lower condom use and therefore increase HIV risk; many were also concerned about the health impact of EC if used regularly. They also stressed the failure rate and held up FP as a safer alternative. EC is currently available over the counter at pharmacies in Burkina Faso, though several facilities (and some pharmacies) do require a prescription. One provider in particular was very worried that promotion of EC might damage the family planning policy of Burkina Faso by detracting from regular contraceptive use; however he was still very much in favour of EC:
I’m saying no to making a policy of promoting it. Making, um, conveying the information to people that it can exist in circumstances, yes!… I never refuse emergency contraceptives to people.
50The broad consensus across both countries seemed to be that providers were in favour of EC being available to a wide range of people through proper health facilities and pharmacies with trained staff on site to give advice on its proper use and warn against the negative effects of frequent use, particularly the risk to women’s health of large hormone doses and the risk of HIV exposure.
51Overall, three responses were identified: those who thought EC should be widely available for anyone; those who thought it should be widely available but through health facilities (because of concerns about health risks, especially HIV), and those who felt its use should be restricted (to certain groups or on prescription only). All three were kept for the typology.
Overall provider acceptance categories
52Once the responses to the typology sub-categories for the “acceptance” dimension were defined, an overall level of acceptance could be ascertained. We ended up with four categories in this dimension:
- Negative (strong concerns over abortion and FP-displacement and restricted access);
- Concern (no abortion concerns but some concern over FP displacement, restricted access);
- Cautious (few abortion or FP concerns; access through health facilities);
- Positive (no abortion or FP concerns; wide access).
2 – Provider responses
53The second dimension of our typology was the nature of provider response to clients presenting in need of EC. Two issues were considered here: first, whether provision of EC was client-initiated or provider-initiated; second, how providers responded to clients presenting with different reasons for having had unprotected sex.
Who initiates EC?
54Although two-thirds of providers in Ghana and Burkina Faso had ever given EC to clients, provision was reactive rather than proactive. The majority of providers in both countries reported giving EC only when the client specifically asked for it:
They come in and tell you, I want Postinor. They know the name very well.
56Another provider reports advising clients over the phone:
58Very few providers in either country explicitly reported proactively recommending EC to clients who presented thinking they were at risk of pregnancy. In such cases, the client was usually a regular FP client already registered with that facility who had missed a pill or had a condom failure.
59Not surprisingly, given the low provider initiation, hardly any providers mentioned that they discussed EC, or pre-emptively promoted it, during the FP consultation. This was probably because EC is not regarded as part of the “family planning method mix” in either country. Although a number of providers in each country referred to EC as “like a contraceptive” or “it’s one of the methods”, the majority wanted EC to be kept as a separate service so that its use could be properly supervised and its regular use discouraged.
60Two clear responses thus emerged: providers who initiated EC themselves (proactive) who were few in number, and those who waited for the client to ask for it (reactive) who were the majority.
How do providers respond to clients?
61Providers indicated that the majority of their EC clients had either had unprotected sex or had experienced an obvious FP method or user failure, such as a condom bursting, or forgetting to take the pill. For clients using FP, providers identified three types of FP failure: method failure (i.e. product deficiency); provider failure (incorrect provision of contraception or inadequate advice) and user failure (taking or using contraception incorrectly). True FP method failure and provider failure are often difficult to detect quickly (unless there is an obvious problem such as condom breakage) as there may be little evidence of failure until the woman suspects that she is pregnant. As a result, women present to the clinics when they are already pregnant – too late for EC. By contrast, people who experience FP user failures or who didn’t use FP at all, generally present earlier because they realize they could be at risk.
62Providers responded differently to different types of presentation, although their perceptions of the clients’ reasons for presenting did not generally appear to be the deciding factor in giving EC or not. Providers generally seemed more judgemental of clients who did not use FP but who walked in to ask for EC. Nevertheless, most providers said they still served such clients, although they may have underreported refusal to serve. User-failure was also seen more negatively, although many providers were sympathetic to clients who were using a FP method (and thus had sound intentions) and experienced a problem – particularly if it was a method or provider failure – “this means that this person really also has a little bad luck from God” (B2), although ironically in these cases, presentation was too late for EC to be administered.
63Overall, three types of responses were identified, regardless of type of client presentation. First, many providers were willing to give EC immediately – either proactively or reactively:
… if they go to a doctor before coming to us then doctor might write it on the prescription that Postinor 2 from the family planning clinic, that is all that there is. But apart from that you don’t need a prescription.
65Second, many providers were willing to give EC, or to refer for it, but did so with certain precautions. Some providers required a prescription from a doctor first but then would serve immediately. Others wanted to check that the client really needed it before giving it and would not give it if they were not satisfied:
There have been a few customers we’ve turned away, we don’t give them.
Q: Why?
We want them to be able to speak with their doctors. We want them to be able to come and speak to us why they want to use it. If they are not willing to talk with us on how they are going to use it and why they are going to use it, I wouldn’t give it to them.
67Others referred the client for EC or gave it directly. For this more cautious approach, the attitude was often
It is a good method to prevent bad things. But the ideal would be contraception itself.
69In the third type of response, a small number of providers were unhappy about providing it and made their views clear to the client, but provided it reluctantly if the client insisted:
What can you say? The product is sold, we cannot refuse to give it.
Overall provision-response categories
71We developed three provision-responses categories:
- Reluctance (reactively responds to client request but reluctantly or inconsistently);
- Precaution (reactively responds to client request but does checks first or refers);
- Immediate (either reactively or proactively responds to client request immediately).
3 – Typology of provider attitudes to emergency contraception
72After defining the categories of our two dimensions, a final typology comprising six types across three categories was developed. It is shown in Table 2. The three categories that emerged were “unfavourable”, “cautious” and “enthusiastic”. The “unfavourable” category included providers who were either completely negative about EC or were very reluctant to dispense it. The “cautious” category included providers who were generally favourable but who had reservations about to whom or how freely EC should be provided, or who were uncertain about promoting it widely because of health concerns. The “enthusiastic” category covered providers who were highly favourable towards EC and wanted to see it widely available, and those who were eager to proactively promote it. Each individual respondent was assigned to a Type within one of the three categories.
Typology of providers’ views and attitudes to EC provision
Typology of providers’ views and attitudes to EC provision
73The next stage was to consider the patterns in background characteristics for the respondents assigned to each Type (Table 3).
Respondents’ characteristics assigned to categories and types
Respondents’ characteristics assigned to categories and types
Note : Each respondent has a single identifier indicating the country concerned (B = Burkina Faso, G = Ghana) and a number.74There were only three providers in the “unfavourable” category (Types 1-2), and no obvious difference between the two countries: one female Burkinabe Catholic midwife (Type 1 – very negative); one male pharmacy assistant in Burkina Faso (Type 2 – reluctant) and one female nurse in Ghana (Type 2 – reluctant but also inconsistent in her responses). Providers in this category had not given EC frequently.
75The two largest categories were the “cautious” category which had 18 providers (6 males, the rest female), mostly from Burkina Faso, and the “enthusiastic” category which had 10 providers (one male, the rest female) mostly from Ghana. Providers in the “cautious” category were mostly mid-ranking professionals (nurses, midwives, pharmacists) with a range of EC provision experience (from a little to a lot). The “cautious” category includes three sub-types. The first (Type 3) comprised a range of providers from both countries who were in favour of giving EC but wanted to control access to it (by age or by education – because of concerns about incorrect use/abuse – or to limit access for frequent users). In Ghana, they were prepared to refer clients for it but not stock it themselves. Many of these providers were pharmacists or pharmacy assistants with moderate-high frequency of giving EC; the three Ghanaian providers who preferred to refer were all midwives with their own delivery clinics who simply didn’t see enough EC clients to make it worth their while to stock it themselves. All providers in Type 4 were from Burkina Faso with moderately frequent experience of EC provision, and all were either doctors or older midwives. They were favourable but cautious about widespread access and active promotion because of a range of concerns. In particular, they feared it would encourage unprotected sex among young people and discourage the use of condoms which protect against HIV as well as pregnancy. In both countries, providers in the three types of “cautious” category expressed some concern over the health impact of frequent repeat EC use. Some providers were cautious simply because they had little or no experience of giving EC. Nevertheless, providers seemed favourable to EC, so long as their concerns could be addressed by provision of EC through qualified health professionals able to give the relevant information and counsel on using a regular FP method.
76Providers in the “enthusiastic” category were from a range of backgrounds, though mostly better qualified and with considerable experience of giving EC. Almost all were from Ghana. There were two types in this category: those broadly in favour of EC who believed it should be widely accessible to anyone who needed it, without restrictions (Type 5); and those who were described as “proactive”. These were providers who, as well as expressing very enthusiastic opinions about EC and its wide accessibility, also said they were proactive themselves in giving and promoting EC (Type 6). There were only three proactive providers, which indicates the work still to be done to allay the concerns of health providers and maximize their willingness to provide and promote EC to clients.
IV – Discussion: is there a place for EC in West African contraceptive programmes?
77Developing a typology of providers’ responses to EC in this study has allowed us to identify the range of attitudes to EC in the capital cities of two sub-Saharan African countries.
1 – Emergency contraception is generally acceptable to providers
78This study found widespread acceptance of the EC pill among providers in both countries for a range of professional reasons including: to reduce unwanted pregnancies; to reduce unsafe or illegal abortions; as an entry point for regular family planning (a view held especially in Ghana); to avoid side-effects of oral contraceptives given as EC. Of particular importance was the finding that few providers had objections to EC on grounds of abortion. The concerns raised in both developed and Southern countries in early studies (Glasier et al., 1996; Senanayake 1996) that EC is abortive and therefore problematic were not borne out by our findings with providers in these two countries, suggesting that understanding has improved in the ten years EC has been available. There are two possible explanations for the differences in degree of favourability between Ghana (generally highly favourable) and Burkina Faso (more cautiously favourable): first, a licensed dedicated product (which is easier to both prescribe and take) has been available for longer in Ghana and exposure to its use is therefore greater; second, Ghana has a more long-standing family planning programme that has increased exposure of women and providers to contraceptive discourse since the 1970s. The picture was somewhat different among the general population, however, and this part of the ECAF study found that there was more widespread concern among both men and women about EC as a potential abortifacient and also as a “chemical” product that may cause sterility (Teixiera et al., 2012).
2 – Concerns about EC replacing regular contraception
79Our findings with providers showed a more mixed reaction to the issue of EC replacing regular contraception, with more respondents expressing some level of concern. Overall, providers in Ghana were less worried, some even regarding EC as an entry point for discussing regular FP use, while more concerns were expressed by providers in Burkina Faso, but mostly related to lack of protection from HIV associated with EC use. The very low prevalence of FP use in Burkina Faso among married women, the main clients of FP clinics, (10% of married women use modern methods versus 17% in Ghana – INSD and ORD Macro 2004; DHS 2009) may also contribute to greater anxiety among providers about any new product with the potential to displace regular FP use. Although the view among many of our respondents supports Marston’s and Moreau’s results that EC availability does not lead to displacement of regular FP in Great Britain and France (Marston et al., 2006; Moreau et al., 2009), some respondents in our study felt that EC did have a legitimate place as an occasional contraceptive for women whose partners were not often home (i.e. travelled a lot). The findings from the general population part of the ECAF study also show that while emergency contraception is usually used as recommended by the medical profession (occasionally and in cases of urgent need), like other post-coital methods, it is also used repeatedly (Teixiera et al., 2012). This would seem to refute Marston and Moreau’s position, and to bear out the concerns of some of our providers, suggesting that the situation in sub-Saharan Africa may be different to that documented in Western settings, though it should be noted that there is no clinical evidence that repeated use of EC has any detrimental health effects.
3 – EC is rarely given proactively
80Providers did not have particularly restrictive views regarding the type of person to whom they were prepared to give EC; they highlighted that EC should be provided to any women experiencing contraceptive failures. This included unmarried and young women and those not using FP, who were all recognized as needing EC although they were regarded in more judgemental terms (see below). The providers’ main reservations seemed to concern young people (under 18), for whom a number thought EC access should be limited. Mixed reasons were given to justify this view, including vulnerability to infection (condoms provide better protection) and a feeling that this age group should remain sexually abstinent. This confirms previous findings that access contraceptives in general is particularly difficult for young people (Stanback and Twum-Baah 2001).
81Despite generally positive in-principle responses by many EC providers across both countries, our study did show, as others have done (Ziebland, 1999; Karasz et al., 2004), a level of caution about actively promoting EC due to a fear of encouraging risk behaviour (unprotected sex giving exposure to STI/HIV infection) associated with a general averseness on moral grounds. Despite studies in Ghana (Lovvorn et al., 2000) and the UK (Marston et al., 2006) showing that unprotected sex does not increase when EC becomes more readily available, the issue of proactive EC provision, including giving supplies to women before they need it, seemed to arouse particular reluctance among providers in our study, as it has among providers in the UK (Fairhurst et al., 2004). As shown in this paper, almost all EC provision was reactive – in response to women who came themselves to actively request EC.
82Providers did appear to make moral judgements about the “worthiness” of clients presenting for EC. Those experiencing a method or provider-related FP failure were regarded with considerable sympathy, since they were seen as responsible, active contraceptors. Those presenting with user failure of a method were sometimes ridiculed as being ignorant or stupid because unable to effectively use a contraceptive, and those presenting without having used FP at all were seen as irresponsible. Other studies have highlighted the dilemmas of the provider-client relationship where medical professionalism sometimes conflicts with particular beliefs or views on morality (Koch and Jones, 2010; Richey 2008), and where providers pass judgement on clients based on their own moral standpoint. This has been found elsewhere in a range of settings (Lupton, 1994; Mokgethi et al., 2006; Richey, 2008), including some situations where providers make a “moral” decision to withhold medical services (Curlin, et al., 2007).
4 – What role for emergency contraception in Western Africa?
83In our study, the reluctance of providers (even those broadly favourable towards EC) to proactively promote EC represents perhaps the most significant challenge to ensuring widespread access to EC. The moral overtones were clearly evident in our study of the general population too, with EC being overwhelmingly perceived as a Northern medical drug which encourages greater sexual freedom for women (sometimes this was seen as positive, usually by women, but more often as negative – introducing ‘non-traditional’ promiscuous relations) and implies a weakening of men’s control over female sexuality (Teixiera et al., 2012).
84Clearly the social context, shaped by the prevailing moral discourse around sexuality, influences both providers and clients/potential clients, as has been documented elsewhere (for example: Lupton, 1994; Mantell et al., 2001; Fairhurst et al., 2004; Richey, 2008; Davidson et al., 2010; Koch and Jones, 2010). Social conservatism may be a barrier to both the uptake of EC by individuals and its proactive promotion by providers, as evidenced by this paper. Education is known to have an impact on FP uptake and this is likely to apply to EC as well once information about the product becomes more widespread. Education can also improve support for, accuracy and timeliness of EC provision among providers, as western-based studies have shown (Harvey et al., 1999; Beckman et al., 2001). The fact that most of the respondents in the “cautious” and (especially) the “enthusiastic” categories tended to be better qualified suggests that training, educational level and provision experience may contribute to favourable attitudes towards EC in our study countries. Even though knowledge of the dedicated EC pill was variable among our providers, our respondents appeared to be eager to know more about it, and several said they had read up on it themselves though had received no formal training.
85Our study is based on a qualitative approach which does not allow for any extrapolation, so our findings cannot be generalized. It is nonetheless important for two reasons. First, it is one of the very few studies of providers’ views towards EC in Sub-Saharan Africa. Second, it has shown that providers in Ghana and Burkina Faso are broadly favourable to EC, which bodes well for expanding its availability.
86Emergency contraception has a place in all family planning programmes, but promoting its use in sub-Saharan African countries raises a number of questions. In many West African countries, including Ghana and Burkina Faso, modern contraceptive use among all women remains stubbornly low, while use of traditional methods remains high, resulting in frequent abortions and high maternal mortality (Ghana DHS, 2009; INSD and ORC Macro, 2004; Harvey et al., 1999; Beckman et al., 2001). In these settings, and particularly in the context of HIV, the promotion of condoms with EC as a backup would have the merit of being a core component of family planning services – especially for those who are not engaged in steady relationships – alongside the usual mix of long-acting methods (particularly injectables) recognized as the most effective means for reducing unintended pregnancies and abortion (Glasier, 2010).
