Door-to-door service delivery has long been the centerpiece of Bangladesh's family planning program. The rural Bangladeshi woman's primary connection with health services has been through the family welfare assistant, a female community-based family planning worker, who visits married women of reproductive age in their homes to promote family planning and supply women with pills and condoms. As of 1995, the Bangladeshi government employed some 24,000 of these personnel, while roughly 6,000-8,000 were employed by nongovernmental organizations (NGOs).¹

The lowest-level government clinic in rural Bangladesh is the family welfare center, which serves some 24,000 people, the average population of a union.* The center is staffed by a male paramedic (known as a medical assistant), who provides basic health services, and a female paramedic (known as a family welfare visitor). These women, who provide family planning and maternal and child health services, typically have a high school education and 18 months of technical training. (Some family welfare centers also have a physician to provide general health services.)

The family welfare visitors are supposed to conduct "satellite clinics" eight times monthly at various sites in each union,^† to give contraceptive injections, insert IUDs, and provide prenatal care. For the most part, however, the system does not encourage women to establish a relationship with a local clinic for reproductive and general health problems as well as for family planning. Instead, the program has relied heavily on the domiciliary services provided by the family welfare assistant.

It is likely that Bangladesh's family planning program will soon become better integrated with the country's primary health care system. As part of the strategy to offer a more comprehensive package of essential health services, the domiciliary delivery system will gradually be deemphasized, and family planning will be provided through static clinics, periodic satellite clinics and supply depots. This reorientation is motivated in part by the perception that a greater emphasis on clinics will make a broader range of reproductive health services accessible to the poor, and in part by concerns over the financial and institutional sustainability of the old door-to-door system. Yet while the current service-delivery mode for family planning in rural Bangladesh is predominantly household-based, some rural women now rely on clinics for these and other reproductive health services. A better understanding of these women's experiences may help guide program leaders in developing strategies for the transition.

This article presents qualitative data on family planning clients' perceptions of health and family planning services. The study was part of a six-year ethnographic research project investigating changing reproductive and gender norms in six Bangladeshi villages and is meant to complement the substantial ongoing research efforts focused on public- and private-sector service delivery in Bangladesh and the impact of these services on demographic and health indicators.² Our interview and observation methods were more open-ended than those typically used in such research in Bangladesh. We quote researchers' direct observations and clients'descriptions to describe through women's eyes and voices problems that need urgent attention if clinics are to become the focal point for reproductive health services in the near future.

Study Setting

The study was based in six rural villages, three in the west-central part of the country, in the Magura and Faridpur districts, and three in the north, in the Rangpur district. According to a late 1994 survey conducted among 838 women in these six villages, contraceptive prevalence in 1993 ranged from 28% to 66%;³ in the aggregate, however, prevalence was comparable to that of rural Bangladesh as a whole—43%.⁴

Although the public-sector services (both general health and family planning) were much the same across the six villages, distances to the nearest fixed facility varied from half a mile to nearly four miles. (At the national level, nearly 50% of Bangladeshi women live within one mile of a family welfare center, according to the 1993-1994 Demographic and Health Survey.⁵) Women in the three west-central villages were much more likely to have visited a fixed clinic, hospital or private doctor's office for services in the past year (31-46%) than were those in the three northern villages (5-20%), according to the early 1994 survey.

The status quo seems to be changing rapidly in Bangladesh, and thus sociocultural constraints should not be a major barrier to the widespread use of clinic-based services within the study communities. For example, when we conducted in-depth interviews in both 1992-1993 and in late 1996 (the period focused on in this article), we observed that over the 3-4-year-interval, women had become more open and articulate about family planning, side effects of methods and sources of services.⁶ Women were also more active in seeking out health and family planning services, both from the door-to-door community workers and from the clinics. The researchers also noticed more mutual support among contraceptive users in the later survey, as women were no longer as reticent about their use of family planning (which some had tried to keep secret in the 1992-1993 interviews). In the 1996 survey, many women reported giving each other information and advice, sharing pill supplies and accompanying one another to clinics. Virtually no one interviewed in 1996 said she had been stigmatized for visiting a family planning clinic.

Data and Methods

The data, collected in late 1996, are taken mainly from two sources. First, interviews were conducted with 34 women in the six study villages who had visited a clinic or satellite clinic for family planning services in the previous three months. Female researchers asked the women about the circumstances of their visit, their experiences with and assessment of the services that they received, and whether they were criticized by their families or others in the village for going there. The interviewers used a guide and also improvised questions.

After eliciting a general account of the clinic visit, the researchers then asked each woman to elaborate on earlier points and to comment on the conditions in the clinic, on how she was treated and how she felt at the time, and on whether in retrospect she was glad that she had gone. The descriptions of clinic visits that we present here are composites constructed from the interviewers' notes of responses to open-ended questions about the circumstances of the visit and the client's opinions about her treatment.

Second, two researchers (one man and one woman) spent a full day observing interactions between clients and staff at the government family planning clinic closest to each of the six study villages, and at the two NGO clinics that were located near the villages. (At one of the NGO clinics, on the day of observation the female paramedic was conducting a satellite clinic in a nearby village, so the observers accompanied her there.)

The observers were asked to record the eight clinics' opening and closing times, the presence of staff, the volume of clients and waiting times to see staff. They were also expected to describe the facility and services being provided (with particular attention to privacy, general orderliness and cleanliness, and existence and condition of latrines for clients). Further, the observers recorded the interactions between clinic staff and clients, commenting on whether the staff treated clients in a respectful manner, whether they seemed to make clients feel at ease, whether they used language clients could understand, whether they responded to clients' needs and concerns, whether they encouraged clients to return, and whether they behaved in ways that reinforced or minimized social distance.

Besides interviews with village women and clinic observations, the article also draws on three other sources of data—interviews with staff of the observed clinics; interviews with community members in each study village, to elicit perceptions about services and to explore how decisions to use these services are made; and interviews (with eight women and with three of their husbands) from selected "focus" households in each village, regarding family planning decision-making, contraceptive use, and contact and experiences with services in the previous 3-4 years.‡ As this intensive study was designed to provide insight into social processes and women's subjective experiences in clinics, the data are not meant to be representative.

Results

Overall Clinic Experiences

The experiences of the 34 recent clients who were interviewed as well as those of women we observed ranged from very positive to very negative. In some instances, the paramedics appeared to make an effort to establish a friendly relationship with their clients. They spoke gently, showed concern, encouraged clients to return to the clinic and assured them that the clinic would help if they had any problems with their method. A woman from Faridpur described her visit to the family welfare center as follows:

"I sat down in a chair inside Apa's^§ office, and she asked me...my name [and] many things about myself—how many children I had, how old my youngest child was, etc....She asked me, 'Have you talked to your husband about it? Does he know that you have come to take an injection ?' ...I asked her, 'Apa, there won't be any problems, will there?' Then she said, 'No, there shouldn't be any major problems. [Paramedic goes on to detail possible side effects.] She told me that if I had any serious problems in the meantime, I should come and see her again....Apa behaved nicely with me. There were no difficulties. ...She examined my eyes. She used an instrument to measure my blood pressure. I didn't suffer any discomfort."

Two others who had visited the same clinic gave similar descriptions—i.e., they received the method they wanted after being screened for contraindications and were treated respectfully and kindly. The fact that women repeated what the paramedic had told them about possible side effects suggests that she is a skilled communicator.

The interviews and observations show that in five of the eight clinics, however, at least some clients were treated harshly by staff. Interactions took on the hierarchical character that is very common in rural areas of Bangladesh, where communication often reflects relationships of political patronage. In poor households, access to land, employment and a variety of other benefits (including many government services) is mediated by more powerful households, and the poor, in turn, must provide a variety of services and display social deference.⁷ Communication tends to be one-way.

In such observed hierarchical interactions, providers seemed to assume that clients had nothing useful to say, and they elicited very little information about the women's problems or histories. When a diagnosis was made, providers simply gave clients pills or a prescription, without telling them their problem. The clients behaved subserviently and did not ask questions. They clearly knew very little about the clinics they attended, such as their hours of operation, qualifications of the staff, what services they were equipped to provide and whether there was a system for determining the sequence in which clients would be seen.

In two of these five clinics, women were observed pleading for services. In one case, the female researcher arrived just before the clinic's opening time—9:00 A.M. The guard opened the door and let her sit down on a bench, where she was joined by a growing number of clients; according to the women who were waiting, the paramedic was habitually late.

When the paramedic finally arrived, at around 11:00 A.M., many clients pushed their way into the consulting room, begging to be seen. She scolded them for crowding around, and some left the room, but she made little effort to establish any system. Over the next two hours, clients continued to enter the consulting room haphazardly, often several at a time. In some cases, the paramedic told the ayah (her helper) to dispense a small amount of medicine or some iron tablets, and then quickly dismissed these women without speaking to them directly.

Other clients were separated out and told to wait. A few women who had come for contraceptive injections or prenatal check-ups seemed anxious. The paramedic spoke gently with them, asking questions to make them feel more at ease. In other cases, however, the paramedic was rude, and scolded her clients. She told a woman who had come because of contraceptive side effects, "When I gave you the pills I told you that you might feel dizzy. Now why are you coming here and complaining, disturbing my work?"

Many women were still waiting when she suddenly shouted: "I'm closing now! My time is 1:00. We have other things to do. Official work!" For the next hour, these women pleaded with the paramedic, who continued scolding them and saying she was closing. Eventually, however, she saw them all.

In another clinic, the paramedic arrived at 11:35 A.M., sat down and was surrounded by clients, who all started to tell her their problems at the same time. She interrupted, saying rather loudly, "You must stop this! Can't you see that there are outsiders present here?" A few women with small crying children were seen first because they were creating such a disturbance.

A woman who had been standing nearby said, "Apa, please examine me." The paramedic replied, "What is wrong with you?" The woman said, "Apa, I am taking the pill. My body is weak. I have no strength and my head spins." The paramedic said, "You should stop the pill and use an IUD!" The client said fearfully, "No, no, I will not use an IUD!" The paramedic then ordered the ayah, "Give her some vitamin pills." The ayah gave the woman seven or eight tablets in a small plastic bag, and the paramedic said, "Take the medicine and hurry up and go! Don't crowd me!"

Later, she refused to examine a woman who wanted to know whether she was pregnant, saying, "No, I will not see any more people now. I have washed my hands. I will not see anyone else!" (It appears that hand washing was done after, not before, patients were examined.) She later relented, however, and told the woman she would examine her.

We interviewed all of the paramedics in these clinics, and they all seemed to know the basic principles of client counseling. However, the contrasts between the paramedics' descriptions of how they treated clients, and our direct observations and interviews with clients, indicate that interpersonal skills, including common courtesy, are applied selectively.

Intrauterine Devices

Women seeking an IUD insertion usually were treated well, but those asking for a removal were not. Only one of 13 women who had requested an IUD insertion** said she had difficulty obtaining one (because she happened to go to the clinic on the day it was closed). Two other women mentioned negative insertion experiences—one said she had an IUD inserted without her knowledge during a menstrual regulation procedure, and the other said the paramedic agreed to perform a menstrual regulation only if the woman consented to use an IUD.

Six out of 10 women seeking a removal^†† experienced difficulty having it done or were treated harshly. One woman's experience illustrates some of the difficulties women encountered in having an IUD removed, especially because the removal was her husband's wish, not hers. When the paramedic asked her problem and why she had come, the client replied that she had a heavy whitish discharge. (She had invented this problem because she suspected that the paramedic might otherwise not agree to a removal.)

The paramedic responded by saying she would give her medicine for this problem and did not examine her. When the husband demanded that the paramedic remove the IUD, she became angry and said, "The IUD cannot be removed until its duration is complete. You had better take the medicine and go."

The woman admitted to the researcher that while she was somewhat upset by the paramedic's behavior, she was secretly pleased at not having the IUD removed. After the couple left the room, however, the ayah suddenly appeared and called them back, urging the paramedic, "Apa, you should remove her IUD! They are the ones who will have a problem. If they have a child, they will have to raise that child. How does it affect you either way?"

This conversation continued for nearly half an hour, according to the client. Finally, the paramedic removed the IUD, but only after saying "Your IUD is very well placed. You don't have any problems with it, so I can't understand why you want to take it out. I tried so hard to persuade you not to remove it, but you wouldn't listen." Afterwards, the woman asked the paramedic if she would be given any medicine. Irritated, the paramedic said, "Go away. You came to have your IUD removed and I've done it. Now why are you asking for medicine?"

The client was not happy with the way she was treated: "If [the paramedic] had been patient and explained things, then perhaps my husband would not have insisted on removing the IUD." She added that she had wanted to find out if she could use a method without her husband's knowledge, but because the paramedic had seemed so angry when they were alone in the examining room, she did not dare ask. She commented, "I was depressed because I had not wanted to have the IUD removed. When I got home, I kept thinking 'Why did I go to the clinic?' If only I'd explained things better to my husband, maybe he would have agreed. Then I would not have had to remove the IUD. They didn't help me or give me any advice. I was too scared to ask them anything."

Several other women had similar difficulties with IUD removals. A Rangpur woman who had had an IUD inserted at home could not get the paramedic to visit her house to remove the IUD. When she got to the clinic, the paramedic was reluctant to remove the IUD, but agreed after the woman lied and said her husband wanted another child. (It was she who wanted another child.) Two women had to go to the family welfare center three times before finally having their IUD removed. In one case, the paramedic was absent the first two times the woman visited; in the other, the paramedic was absent once and on the second visit, the client had succumbed to the paramedic's pressure to keep the IUD despite side effects.

One NGO client we observed complained of pain in her lower abdomen. She told the paramedic that she suspected the paramedic in the nearby government clinic had only pretended to remove her IUD—when she had asked to see the device, the paramedic refused to show it to her.

The fact that women are afraid to switch to the IUD despite ongoing problems with other methods, that they feel compelled to lie to convince paramedics to remove it and that they suspect providers only pretended to remove the device suggests a widespread perception (or suspicion) that the family planning program is more interested in getting women to use the IUD than in serving their true needs.

Medication

Medicine—including prescription and nonprescription drugs as well as nutritional supplements—was a constantly recurring theme in both interviews with clients and in the exchanges observed in the clinics. In addition to contraceptives and vaccines for immunization, the family welfare centers stock basic drugs and nutritional supplements for treatment of common illnesses, with a particular focus on the health problems of reproductive-age women and of infants and children (such as intestinal diseases, anemia, respiratory infections, and eye and skin infections). These are supposed to be provided free of charge, as are services such as menstrual regulation and IUD insertions and removals. (With some exceptions, NGO family planning clinics offer a more limited range of health services, and these clinics charge a small amount for contraceptives and drugs.)

Many, if not most, of the clinic interactions we studied revolved around a request for medication; this was often the primary reason for clinic attendance and frequently became a focal point for bad feeling and suspicion between providers and clients. Many clients in public clinics believe that staff withhold medicine (and sell it illegally outside the clinic system); the clinic staff see their poor clients as constantly badgering them for medicine to which they are not entitled.^‡‡ As one paramedic was observed scolding a client, "You don't have any problems at all. You only come to the clinic because you get medicine every time you come here." The paramedic then gave the woman roughly 30 iron tablets, without examining her.

One woman from Magura, who had been receiving contraceptive injections through satellite clinics in her village, was having problems with prolonged menstrual bleeding. One day she went to the family welfare center to seek help. After waiting for some time, she explained her problem and the paramedic suggested that she use another method, such as the pill or the IUD. The woman was afraid of trying either of these methods, however, and wanted vitamin tablets and asked for them two or three times. As she explained to the researcher:

"Before, when I had problems with my periods, I went to [the family welfare center] and got some vitamin tablets. When I took those, my periods would stop. This time, they said that they didn't have any vitamins, but I saw that the cupboards were full of medicine. The woman opened one of the containers and gave me a few pills. When I went forward to take a closer look, the daktar scolded me and said, 'Why are you going there?' I said, 'Please give me a few more pills.' She said, 'You can't have any more today.'...The daktars are sometimes nice, but not when you ask for medicine."

In cases where a woman's family perceives she needs treatment badly, often the husband or another relative goes to a drug seller, doctor or clinic to obtain medicine, rather than arrange for the woman to be seen by a health care provider. Thus, the tendency for health personnel to often hand out vitamins and medication without examining patients or taking a thorough history might discourage women from seeking care in person.

Acceptance of Clinics' Shortcomings

In general, the clients interviewed in this study were not very critical of the condition of the clinics, nor of the quality of the services they received, and they assumed the paramedics were technically competent. They appreciated even the smallest gestures of concern and courtesy, and criticized staff only when their behavior was unusually rude. We learned of few cases in which female clients challenged a paramedic or openly protested rude treatment: One woman noted that another client had argued with a paramedic over medicine, and one woman refused an examination out of embarrassment.

In our direct observations, we found all of the clinics to be unacceptably dirty. The clients, however, simply did not notice the lack of hygiene and commented negatively in only one extreme case (a government clinic in Magura).

One NGO clinic that we visited had dirt floors and tiny cramped rooms, partitioned only by filthy curtains and improvised walls made of dusty, disintegrating woven mats. There was no waiting area to speak of. When asked specifically, one IUD removal client admitted that the clinic was dusty, was not very clean and lacked water. But she added:

"We go there anyway because Apa treats us very nicely. She takes much care in giving us treatment....I got the medicine and treatment that I needed in the clinic. I didn't have to worry about anything. If we go to the government clinic, the apas always tell us to go away because they have no medicine. They say they will remove the IUD only when its duration expires....Besides, going to the clinic gives me a chance to visit places outside."

In most of the clinics that we observed, there was visual privacy in the examining room, but the consultations could usually be overheard, and often other patients would crowd into the paramedic's room. No one complained about this, although one woman said that she used to feel embarrassed but now was used to it. None of the women reported being alarmed when paramedics failed to wash their hands before performing examinations, or when they used the same pair of gloves for pelvic exams or IUD insertions on several clients. These clients probably believed that sterile gloves are intended only to protect the paramedic.

Unless specifically asked, no one mentioned that the latrine was filthy, or that it was kept clean but locked and designated for staff and IUD clients only. Although some women reported rudeness, and a few were intimidated by the providers' harshness and hesitated to ask them questions, clients usually felt that the providers were skilled and said they would return to the clinic.

Two members of the research group systematically reviewed the transcripts from the interviews with recent clients to rate each client's overall experience as "positive," "negative" or "mixed." These ratings were then compared to the clients' own responses to questions asked at the end of the interview on their overall satisfaction with the clinic.

The researchers' ratings of the clients' experiences, on balance, were more negative and mixed than positive. For the 30 women for whom detailed transcripts allowed an overall rating, one researcher rated 11 women's experience as positive, six as mixed and 13 as negative; the other researcher, reviewing the same transcripts, was somewhat more critical, assessing the experiences of seven women as positive, 12 as mixed and 11 as negative. The researchers' negative assessments were based mostly on evidence in the transcripts of the provider's rudeness and failure to communicate with the client; we also gave negative ratings when paramedics refused to perform menstrual regulation without payment, despite the client's pleading, and when a paramedic's tardiness resulted in extremely long waits.

The clients themselves were less critical about their treatment: Nearly half of the 28 women who commented on their visit indicated that their overall experience was good (13), saying in most cases that the paramedic at the clinic was skilled in her work, that she had helped them and that they were glad they had gone to the clinic. The ratings of the other half of the sample were divided more or less equally between mixed (8) and negative (7), and among these 15 women, only two said they would not go back to the clinic. Most expressed a willingness to overlook rude behavior and various discomforts, as long as the outcome of the visit was positive. For example, a woman who had reported being too intimidated to ask questions because the paramedic seemed so irritable said:

"I wasn't very satisfied with the treatment that she provided....[But] if I don't use some kind of contraceptive, I will have a baby. We are poor people. How would we feed another child, arrange for its marriage? I am very fortunate to be able to get the family planning injections."

Another woman waited almost two hours in a hot, uncomfortable room with no fan. The clinic was dirty and littered with trash. She was thirsty, but the tap was broken. The ayah was unfriendly, and the paramedic belittled her and told her to stop coming around begging for medicine. Yet she explained:

"Even though they behaved badly, I have to be content. We are lucky if we can get the free medicine that they give out at the clinic. If I have another problem, I will go to the clinic again, because they don't charge a fee and you can get free medicine. We are poor people; how are we going to get better treatment than this?"

Conclusions

Current Problems

The issues that emerge from our research have more to do with power relations between clients and providers, accountability and broad institutional policies than with technical skills, standards and protocols. Three themes surfaced repeatedly.

•Communication problems. Hierarchical modes of interaction and poor communication were evident in many of the encounters between service providers and clients. The more egalitarian style prevalent in women's interactions with domiciliary workers⁸ raises the question of whether less-hierarchical client-provider interactions might be "imported" into the family welfare centers by giving community-based family planning workers a greater role in these clinics.

Certainly, one cannot assume that less-hierarchical communication would occur automatically: Patterns of interaction that express power and status differences are not easily altered when underlying social inequalities remain unchanged. Nonetheless, behavior is not immutable. Giving the family welfare assistants greater roles and status in clinics could be one of several strategies for making client-provider interactions more constructive and beneficial for clients.

•Conflict over medication. The issue of entitlement to medication often sparked bad feeling and suspicion among clients, who believed that staff were withholding and illegally selling drugs. On their part, the clinic staff seemed to see clients as constantly pestering them for medicine to which the women were not entitled.

Public-sector clients often do not receive the medicine they believe they need, for a combination of reasons. In some cases, the medicine they request (or hope to receive) may be inappropriate for the problem. In others, clinics may have run out of specific medications and nutritional supplements because of genuine shortages in the government health system or inadequate logistics. Drugs are also probably sold illegally outside of the clinics, although the magnitude of this practice is difficult to assess. In any case, service providers appear to respond to drug shortages with informal rationing, which creates tension between providers and clients. Even when drugs are not being sold illegally outside the clinic system, shortages may create the suspicion that they are.

To improve the quality of care offered in clinics, a way needs to be found to shift the emphasis away from simply distributing medicine and vitamins, and toward examining, diagnosing and counseling clients. Accountability mechanisms are needed to inhibit the illegal sale of drugs in public-sector health facilities and to dispel the popular perception that this is taking place.

•Uneven application of counseling skills and courtesy. The contrast between how several paramedics responded in their interviews and how they acted when they were observed suggests that rudeness to clients is more than a reflection of ignorance, and that interpersonal skills and common courtesy are sometimes used quite selectively.

Most of the paramedics we encountered seemed to know how to treat clients well, implying that domineering behavior, while normal, need not be inevitable. Although they had been taught that establishing communication and even being courteous was part of the job, institutional policies and incentive systems seem to convey the message that only certain types of clients deserve decent treatment. Clinic staff need to hear from the institutions that employ them that they are there to serve all women equitably. This concept probably cannot be communicated through a single training course, but must be embedded in broader institutional policies, management systems, norms and incentives.

Improving Quality of Care

Efforts to improve the quality of care in family planning programs have tended to concentrate on Africa and Latin America. The exceptions in Asia have been method-specific quality improvement efforts, and possibly the Matlab project in Bangladesh.⁹ The lesser emphasis on quality of care in Asia most likely results from the predominance of large public-sector programs, like that of Bangladesh, which historically have been driven by demographic objectives and organized around method-specific targets. In Bangladesh, the training that both government and NGO paramedics receive on counseling and client relations might be overshadowed by the more constantly reinforced message that their job is to increase the use of clinical methods.

The voices and views of clients are considered indispensable in efforts to improve the quality of care in family planning and reproductive health programs.¹⁰ To hear women's voices, one must ask questions in ways that encourage them to speak; furthermore, to understand what women are saying, nonarticulated aspects of their lives often have to be taken into account as well. The differences in the researchers' and the clients' overall assessments of care in this study raise several points about what hearing women's voices means in the context of research on quality of care. The different perspectives reflect different standards and concepts of health care, as well as the more limited experience and options of the rural women, many of whom had never experienced any other standards of care.

On the other hand, these women obviously recognized and disliked the rude behavior of the clinic staff. Those whose problems were not alleviated felt frustrated, but they went to the clinic only because they had nowhere else to turn.

Standard research approaches have provided at best a very limited picture of the perspectives and experiences of actual and potential clients,¹¹ despite the impressive array of "client-centered" research and management tools developed to identify and address quality-of-care problems.¹² Even with the increased application of client-centered management approaches, clients often feel like outsiders in health-care systems and family planning clinics.

We believe that the approach of our study, which employed well-trained local researchers who were relatively unfamiliar with the structure and norms of a family planning program or clinic facility, is well suited for client-perspective research; such outside researchers are better able than insiders to appreciate the confusion and discomfort of the client. They may also be more likely to consider extrainstitutional factors, such as the political and social objectives and environments in which services are provided, and to suggest new ways of looking at problems.

According to the most frequent comments and complaints of the women with whom we spoke, improvements in quality of care would almost certainly include better, and free, access to effective medicines to alleviate health problems and side effects. This suggests that women's needs would be more effectively met if family planning services were better integrated into health services. It also underscores the need to address the difficult issue of corruption, which breeds mistrust on the part of clients and cynicism on the part of service providers, many of whom are simply caught in the system.

A new approach to providing information and education is needed so that rural people understand basic reproductive health, know their rights and what to expect, and make informed decisions in utilizing health services.^§§ People who are better informed about public-sector health and family planning services would also be in a better position to demand that service providers become more accountable to their clients.

Many NGOs in Bangladesh are involved in organizing and mobilizing the poor, and especially women, in poverty-alleviation and health-promotion programs. Pressuring the health system to become more conscientious and accountable would be consistent with the philosophies and mandates of many of these organizations, but at present relatively few are doing so, because they fear that such activity might be perceived as political, and might lead to confrontation with the government. In the future, donors might provide funding to encourage and assist NGOs to successfully pressure the health system, by developing health education and advocacy skills among the rural poor.