Dept. of Population and Family Health Sciences
Johns Hopkins University, Baltimore, MD 21205
Gillian H.C. Foo
Independent Consultant, Baltimore, MD
Dept. of Population and Family Health Sciences
Johns Hopkins University, Baltimore, MD 21205
In recent years, there has been growing recognition among researchers and policy makers that the quality of care provided by family planning programs in developing countries is an important determinant of the overall success of such programs, and a central factor in whether services are fully and effectively utilised by the population they are designed to serve. Quality services are also considered to be integral to the overall reproductive health needs of individual clients.
The issue of quality of care has, in recent years, become increasingly salient within India. Despite more than four decades of effort, the Indian program remains characterised by modest achievement and considerable unfulfilled promise. While recognising that fertility behavior is shaped by a range of both demand and supply factors, there has been a growing consensus among researchers, policy makers, and observers that the program itself—in terms of program philosophy, priorities, and the actual implementation of the service program—must be given a primary explanatory role for the limited success of the program to date (Conly and Camp, 1992; Measham and Heaver, 1996).
A starting point in formulating policy and programmatic interventions for improving the quality of services is a detailed and realistic understanding of existing standards of care within the Indian family planning program. As one of the oldest and most ambitious efforts to influence human fertility in a highly significant country demographically, the Indian program has been closely followed and analyzed over the four decades of its existence. A considerable amount has been published on the Indian program. Much of this work has focused upon population policy within India-- historically (Narayan and Kantner, 1992; Visaria and Chari, 1998), during the politically turbulent 1976-77 Emergency period (Gwatkin, 1979; Kocher, 1980; Pai Panandiker and Umashankar, 1994), as well as during the most recent period which has been characterized by a major shift in policy toward a reproductive health approach (Visaria, Jejeebhoy, and Merrick, 1997; Visaria and Chari, 1998). There has also been much work on the structure, organization, and financing of the Indian family planning program (Narayan and Kantner, 1992; Berman and Khan, 1993). And finally, there has also been considerable attention devoted to the outputs of the Indian program, in terms of contraceptive prevalence and method mix (Soni, 1983; Satia and Jejeebhoy, 1991; IIPS, 1995).
Despite this extensive body of work, information on the actual quality of services at the level of implementation—particularly at the interface between the service program and its clients—remains surprisingly limited. Much of the existing evidence on quality of care within the Indian program has only recently become available, and this information for the most part remains highly fragmentary, generally unpublished, and therefore largely inaccessible to the growing research and policy communities interested in this issue. In the absence of detailed information on the quality of services, policy discussions on this issue have tended to remain at a very general level and concrete recommendations for improving the quality of services elusive.
Our objective in this chapter is to review and synthesize recent empirical evidence on access and quality of care within the Indian family planning program. In integrating such a broad and diverse body of evidence, we draw heavily upon the quality of care framework articulated by Judith Bruce. Building on the work of Donabedien (1988), Bruce outlines six elements—choice of methods, information given to clients, technical competence, interpersonal relations, mechanisms to encourage continuity, and appropriate constellation of services—which collectively capture the multi-faceted dimension of quality of care. In the following sections, we first consider evidence on access to and availability of services, before reviewing data on the various quality of care elements identified by Bruce.
Accessibility and availability of services
Within the Indian government program, each Auxiliary Nurse Midwife (ANM), the key multipurpose extension health worker, is expected to visit every household within his or her work area (typically consisting of 5,000-6,000 households) at least once in two months to provide family welfare services. These include disease surveillance, information on and services related to maternal and child health, as well as motivation for family planning and the provision of some contraceptive methods. Almost all studies indicate significant shortcomings in worker-client outreach effort—in the frequency and regularity of outreach visits, in the time devoted by workers to outreach activities, and in the duration of time spent with clients during outreach visits.
The available community-based evidence suggests considerable variability in the level of outreach visitation by ANMs, largely by geographical location, with significantly higher levels of reported outreach in the South and Western than North Indian states. In the four-state study by Roy and Verma (1999), 89 percent and 93 percent of women in Tamil Nadu and Karnataka, respectively, report having been visited by a female paramedical worker within the last three months. An interesting exception to this pattern of regional variation in outreach is Kerala, where an earlier Indian Council of Medical Research study found that only 40 percent of women reported having been visited by a family planning worker during the preceding six months (ICMR, 1988). An explanation may lie in the already very high levels of contraceptive use and the ready access to clinic facilities by most women, thus requiring less outreach effort by service providers. Worker outreach visitation rates in the Western Indian states have also been quite high. Two studies from rural Gujarat found that 96 percent and 73 percent of women reported having received an outreach visit by a female health worker in the past six months (ICMR, 1988; Visaria, 1999). The study by Murthy (1999) in rural Maharashtra, in contrast, found that only 51 percent of women had been contacted by a female outreach worker within the past three months.
Studies of North Indian states have generally reported much lower levels of worker outreach. At the extreme are findings from Uttar Pradesh, where less than 10 percent of women interviewed reported having been visited by a female outreach worker during the preceding three months; household visitation rates by male health workers were even lower (SIFPSA, 1996; Khan et al. 1999a). This very limited level of program outreach does not appear to have changed significantly over the past two decades (Misra, et al., 1976). Other studies of program outreach effort also report very low levels of female worker-client contact in the North Indian states--under 10 percent in the past year in Madhya Pradesh (Talwar, 1988), and in the past six months in Bihar (ICMR, 1988).
Not all studies from North India, however, report such low levels of outreach. A study in Bihar found that 34 percent of women respondents had received a visit by health/family planning staff (primarily ANMs) during the preceding three-month period (Khan et al. 1994).A second important exception is the four-state study by Roy and Verma (1999), which found that 53 percent of respondents in Bihar and 61 percent of respondents in West Bengal reported a home visit by a female worker during the preceding six months. These findings suggest that even within the North Indian region, there may be considerable variability in terms of outreach efforts and visits.
There is also limited evidence of selectivity in worker visits according to the location of the community. The previously cited study in Maharashtra found that respondents residing in villages more remote from villages in which health facilities were located--that is, where the ANM was assigned--were much less likely to have reported a recent visit by a health worker, to have been visited for meaningful lengths of time, and to have received other maternal and child health services (Murthy, 1999). An earlier study in Uttar Pradesh also found a much greater tendency for workers to visit communities and households which were accessible to main roads (Blaikie, 1975). A more recent study from the same state, while finding universally low levels of outreach reported by all respondents, found that women residing in more remote villages were least likely to be contacted (Khan et al. 1999a). A 1987-89 study of a nationally-representative sample of clinics by the Indian Council of Medical Research found that more remote villages--e.g., those without a primary health center or subcenter-- were more likely to be characterized by a complete absence of IUD and sterilization acceptors (ICMR, 1991). Qualitative data from Madhya Pradesh support the observation that ANMs’ visits to outreach areas are irregular. In this study, villagers from a community in one ANM’s work area made it clear to researchers that neither the ANM nor her male counterpart had visited their village in the past six months and thus had little credibility within the community (Barge and Lakshmi R. 1999).
A central factor in the low levels of outreach effort is the limited time spent by workers on their job. Time-use studies from India have generally found that female workers spend far less than the mandated time on the job. A time-use study from Karnataka found that workers spent on average four hours per day on the job, of which only 68 percent of time was actually devoted to service delivery (Bhatia, 1999). Other time-use studies, however, in Maharashtra (Iyer and Jesani, 1999) and in Madhya Pradesh (Barge and Lakshmi R., 1999), have reported slightly longer workdays and greater effort devoted to outreach activities, although the possibility exists that workers may have inflated their performance given the relatively short observation periods in both studies. Considering these work patterns, it is not surprising that workers generally spend little time with clients during their outreach visits. A study in Bihar reported that 41 percent of respondents felt that the amount of time spent by the fieldworker with them was very short; only 31 percent were fully satisfied with the time spent by the fieldworker (Khan et al. 1994). In a study in Maharashtra, almost two-thirds of respondents reported that the ANM had spent less than five minutes during the most recent household visit (Murthy, 1999).
Studies of government health facilities have, in contrast, generally shown relatively high rates of utilization. In the previously cited four-state study, a very high percentage of respondents--ranging from 58 percent in West Bengal to 74 percent in Karnataka-- reported having visited a primary health center or subcenter during the preceding six months (Roy and Verma, 1999). Other studies have also reported significant rates of clinic usage among respondents and their families. In general, clients perceive private sector health and family planning services as superior to those offered by the government program, as clearly illustrated in Murthy’s findings whereby respondents consistently rank specific dimensions of service quality higher in the private sector compared to the public (Murthy, 1999). As Ravindran (1999) has suggested, the shortage of women doctors in the public sector clinics may also draw more women clients to private clinics where women doctors are more likely to be found. Nevertheless, government clinics continue to be utilized--substandard service notwithstanding—in part, because the monetary costs to clients are minimal.
The official policy of the Indian family planning program is that clients should be able to voluntarily choose a contraceptive method (either terminal or temporary) from the full range of methods available, and that they should be provided full information on these methods. In practice, this appears to rarely occur. Almost all studies from India on this issue have found that clients generally receive very limited information from providers in the way of method choice. In Maharashtra, over 60 percent of respondents interviewed indicated that during visits by ANMs they were not told about spacing methods. A study in Bihar found that only 40 percent of respondents were informed about more than one available contraceptive choice (Khan et al. 1994). Other studies have found a similar pattern of limited contraceptive choice to clients, with options focused upon female sterilization, and to a lesser extent the IUD (Khan, et al. 1999a; Ravindran, 1999). Almost all studies have found that male sterilization is rarely offered as an option (Roy and Verma, 1999: Bhatia, 1999; Khan, et al. 1999b). At the same time, providers argue that shortages and erratic supply of temporary contraceptive methods make it unrealistic for them to offer clients a choice of methods. In Uttar Pradesh, for example, IUDs were reported to be regularly available at only 50 percent of the health centers visited, and one third of such centers reported that oral pills were not regularly supplied (Khan et al. 1999a).
A somewhat more optimistic picture is presented by the four-state study of Roy and Verma (1999). A strong emphasis by clinic staff upon sterilization was evident, especially in the South Indian states of Tamil Nadu and Karnataka, where 58 and 56 percent of respondents reported that this method was always emphasized. Interestingly, about one-half of respondents in these two states indicated that clinic staff also always emphasized spacing methods of family planning, indicating some effort at providing clients with contraceptive options. With respect to outreach visits by ANMS, between 21 percent (Karnataka) to 39 percent (West Bengal) of respondents reported that the ANM emphasized only sterilization; slightly smaller but significant percentages of respondents reported that the ANM discussed both sterilization and spacing methods. Only in Karnataka did higher percentages of ANMs appear to place emphasis upon both types of methods (43 percent). Khan et al (1999a) found in their survey from Uttar Pradesh that 63 percent of respondents reported that the worker who visited them insisted that they adopt a specific method; in 71 percent of such cases, the method was female sterilization.
Data on prior contraceptive use further substantiate the restricted patterns of contraceptive method use—and indirectly the absence of choice-- which prevail within India. The 1992-93 National Family Health Survey reported that 82 percent of sterilization acceptors nationally had used no other method prior to undergoing this procedure (IIPS, 1995). Similarly, a study in Gujarat found that only five percent of sterilized women had been offered a choice of method other than sterilization; similarly, only 12 percent of IUD acceptors said they had received information on any other method prior to acceptance (Visaria, 1999). Comparable findings were reported from a study in Orissa, where only 14 percent of family planning acceptors (largely sterilization) were informed of the availability of other methods (Khan et al. 1990).
Studies of service providers present corroborative evidence that little emphasis is given to client preferences for method choice. Verma and Roy report that 93 percent of ANMs interviewed in Karnataka stated that it is the provider who decides which method the client should use; in the other three states, approximately 75 percent of the ANMs held this opinion (Verma and Roy, 1999). These findings assume added importance given the consistent findings of a distinct stress placed by providers upon a single method, female sterilization.
A 1987-89 Indian Council of Medical Research study noted a strong emphasis by ANMs upon sterilization, with less regard for the IUD, and little emphasis upon either oral pills or condoms (ICMR, 1991). An earlier three-state ICMR study also found that workers themselves reported a very strong emphasis upon tubectomy for their clients; ranging from 63 percent in Gujarat, to 79 percent in Bihar, to as high as 86 percent in Kerala. Emphasis by workers upon the IUD was minimal, with the exception of Kerala, where 30 percent of workers reported also emphasizing this method (ICMR, 1988). Two qualitative studies in Madhya Pradesh both found a pronounced tendency by workers to emphasize sterilization at the exclusion of other potential methods, primarily in response to intense pressure for family planning targets (Singh and Kumar, 1988; Barge and Lakshmi R., 1999). Focus group discussions conducted with ANMs in Uttar Pradesh revealed that providers promote a specific method according to the parity of the woman and the gender composition of her children and, as such, believe that they are promoting method-mix (Khan et al 1999b).
There is some evidence to suggest that program staff make selective decisions about the provision of contraceptive choice and information. Visaria and Visaria (1992), in their study in rural Gujarat, observe that: "… It almost appears as though FP planners decide in advance what is best for individual couples." (p. 129). These choices may be influenced by both the location and socioeconomic characteristics of the client. The study by Murthy (1999) in rural Maharashtra, for example, found that women residing in more remote communities and also women who were less educated were significantly less likely to have been informed about both spacing methods of contraception and method side effects.
At the extreme, there is some evidence of involuntary choice of contraception by some Indian public sector providers. In-depth studies in Tamil Nadu and Uttar Pradesh cite examples whereby a woman’s access to abortion has been made provisional upon her agreeing to undergo a sterilization (Khan et al. 1999; Ravindran, 1999). The same study in Tamil Nadu also reported cases where women had received an IUD after delivering without their knowledge or consent (Ravindran, 1993). A second, more recent study in Tamil Nadu also found that the practice of involuntary insertion of IUDs in women immediately after delivery appeared to be both condoned by program staff and a common practice in at least some large urban government hospitals (Van Hollen, 1998). While it is difficult to gauge how extensive these practices are, since many women willingly adopt long-term contraception post-delivery. The Tamil Nadu study reported that a majority of women who delivered in urban government hospitals appeared to have received a method either unknowingly or without their explicit approval (Van Hollen, 1998). Several other studies have also reported that access to abortion services is often conditional upon acceptance of a long-term contraceptive method (Gupte, 1997; Ganatra, et al. 1998; Khan, et al. 1999b), suggesting that this practice may be very widespread in India.
Information to Clients
Available evidence from India indicates that information given to clients by providers on family planning methods is frequently inadequate, that side effects are often not clearly delineated, and that clients are not fully or effectively counselled in how to deal with them. Some of the most detailed information comes from the four-state study conducted by the researchers at the International Institute for Population Sciences (Roy and Verma, 1999; Verma and Roy, 1999). Almost one-half of respondents in the surveys of the North Indian states of Bihar and West Bengal reported that ANMs rarely (if ever) discussed contraceptive side effects during their visits. Exit interviews with family planning acceptors revealed a similar picture. Slightly more than half of acceptors in the states of West Bengal and Tamil Nadu were informed how the method worked and should be used; fewer than half were provided information on side effects and their management (Roy and Verma, 1999). A somewhat more favorable picture was evident among the small number of acceptors interviewed in Bihar and Karnataka.
Interviews with service providers revealed a generally similar picture. Only a minority of service providers indicated that they discussed with clients their reproductive goals—17 percent in West Bengal, 21 percent in Bihar, and 36 percent in Tamil Nadu. Only in Karnataka did a majority (55 percent) of service providers discuss reproductive goals. Similar percentages were evident with respect to providers’ discussions of method side effects (Verma and Roy, 1999).
In a survey in Uttar Pradesh, among clients contacted about family planning, a majority were informed solely about the advantages of specific methods; only about one-quarter were given information about both the advantages and disadvantages (Khan et al. 1999a). Similar shortcomings in the extent to which information was provided to clients were observed in studies in Orissa and Bihar (Khan, et al. 1990; Khan et al. 1994). Qualitative data from Kerala and Tamil Nadu support survey findings on the incomplete nature of information on contraceptive methods received by family planning users from providers (Ravindran, 1993; Ramanathan, 1995). Providers tend to highlight a contraceptive’s effectiveness and avoid raising other relevant issues which may be less positive, such as contraindications, so as not to deter a potential acceptor. A focus group study in Uttar Pradesh revealed that clients frequently received only limited information on contraceptive side effects, and respondents believed that workers deliberately withheld such information because of fears of deterring them from accepting contraception (Levine et al., 1992). A study of ANMs in Uttar Pradesh found that while they were very knowledgeable when interviewed about different contraceptive methods, how they work, and their side effects, they conveyed very little of this information to their clients (Khan et al. 1999b). Clients were not told about the side effects of the contraceptives and were only asked to return if they had a problem.
The extent to which information is provided to clients in sterilization camp settings appears to be no better, and possibly worse, than in other contexts. Studies in a wide range of settings within India document the almost complete absence of both pre-operative and post-operative counselling which occurs within these settings (Ramanathan et al. 1995; Mavalankar and Sharma, 1999; Lakshmi R. and Barge, 1999), missing a major opportunity for educating clients about the sterilization procedure, potential complications, as well as other reproductive health concerns.
A number of quantitative and qualitative studies have attempted to assess the nature of interpersonal dynamics between service providers and clients, as well as clients’ perceptions of such exchanges. Some of the earliest evidence on this issue comes from a study of client-provider interactions in Uttar Pradesh, which found considerable variability in the interpersonal skills of family planning workers, and the extent to which they were able to establish rapport with rural villagers (Rao, 1977). In the four-state survey by Roy and Verma (1999), providers’ interpersonal behavior in Bihar and West Bengal received distinctly lower ratings by client respondents than in Karnataka and Tamil Nadu (Roy and Verma, 1999). Respondents in Bihar and West Bengal were much more likely to express negative views on such issues as whether clinic staff and the ANM paid attention to their health and family planning needs and generated confidence to adopt/continue contraceptive use. A similar picture was obtained from a separate study in Bihar, in which a majority of respondents expressed lukewarm or negative views about their experience with government outreach workers on a range of dimensions (dependability, sympathetic to client needs, responsive to questions), and only a minority rated workers favorably on these aspects (Khan et al. 1994).
In contrast, respondents from the South Indian states of Tamil Nadu and Karnataka expressed largely positive evaluations on almost all dimensions of interpersonal relations. Similarly, exit interviews with clients revealed consistent differences, with a majority of South Indian respondents, but a minority of North Indian respondents, expressing positive views on dimensions such as the cordiality of clinic staff and the adequacy of privacy (Roy and Verma, 1999). Roughly three-quarters of respondents in rural Maharashtra indicated satisfaction with the following aspects of government services-- whether staff at government clinics were friendly, whether providers and staff were attentive, and whether their questions were satisfactorily answered. However, private sector services received consistently higher ratings than public sector services on all of these dimensions (Murthy, 1999).
The available qualitative data present a mixed picture on interpersonal dynamics. Direct observations of client-provider interactions have tended to be positive, with researchers generally reporting empathetic and patient care on the part of the provider (Barge and Lakshmi R., 1999; Ramanathan, et al. 1995). In contrast, a qualitative study in Tamil Nadu paints a decidedly negative view of clients’ encounters with support staff and providers in health facilities: the observations are overwhelmingly negative, citing verbal abuse by staff and nurses, and demands for payment (in an ostensibly free system) before rendering even the most basic services (Ravindran, 1999). While it remains unclear how widespread such treatment is, observation studies of provider-client interactions in clinics in New Delhi and Tamil Nadu vividly illustrate the frequently harsh and derogatory treatment which poor Indian women experience when seeking family planning services within the public sector (Gupta, 1993; Nataraj, 1994; Ganatra, et al. 1998). A focus group study in Uttar Pradesh also documents similar perceptions among women respondents that the staff and medical officers in government clinics are often rude and discourteous to clients (Levine, et al. 1992).
Within sterilization camp settings, the available evidence suggests that empathy with clients is largely absent, and that little consideration is given to the dimension of interpersonal relations. Patients appear to be treated with little respect or sympathy and are not provided with the most basic information which could make this a less anxiety-provoking experience (Mavalankar and Sharma, 1999; Lakshmi R. and Barge, 1999; Parveen, 1995; Ramanathan et al. 1995). Observations from these studies indicate that nurses and doctors are intolerant of complaints from sterilization patients and respond to them impatiently and harshly. Little is done to ensure the physical comfort and safety of patients. The modesty of women patients is not respected, with the lack of spatial privacy at these camps, the inability to provide surgical gowns, and the practice of male operating theater staff assisting women patients into the lithotomy position, when they are most exposed.
Technical Quality of Care
Important indicators of technical quality of care include full and accurate knowledge by providers, as well as acceptable clinical practice, related to family planning and reproductive health service delivery. Studies from India have pointed to significant lapses among providers in both of these areas. In their four-state study of providers, Verma and Roy (1999) found ANMs’ knowledge on these issues to be quite variable. A relatively high percentage of ANMs knew the gestation period up to when to recommend that a client can safely seek an abortion--over 90 percent in the South Indian states of Karnataka and Tamil Nadu and over 70 percent in West Bengal and Bihar. Knowledge of basic issues such as what a tubectomy procedure is was somewhat lower—ranging from 68 percent in Karnataka to 90 percent in Tamil Nadu, and the proportion of workers who correctly understood when the fertile period occurs in the menstrual cycle was even lower—ranging from only 39 percent of workers in Bihar to 86 percent in Tamil Nadu. An earlier study in Madhya Pradesh reported that over one-half of female workers did not understand how to take blood pressure, and over three-fourths lacked knowledge of many basic pathological tests (as cited in Satia and Giridhar, 1991). The study by Visaria (1999) in Gujarat demonstrates the prevalence of inaccurate information among a substantial proportion of ANMs concerning contraindications for the provision of oral contraceptives to potential clients. Studies also underscore significant gaps in providers’ actual clinical practices. In a study of IUD acceptors in Gujarat, for example, only 49 percent of acceptors reported receiving a check-up before the IUD insertion, and only 41 percent reported that the provider washed her hands or put on gloves before inserting the IUD (Visaria, 1999).
Much of the evidence on technical quality of care in India is obtained from studies of sterilization services provided through mass programs or ‘camps’, which remain a key source for sterilization throughout much of rural India. Townsend et al. (1999) define a camp as "...any grouping of patients for a specific service, independent of the site, or quality", and this approach is used to maximize the number of women on whom the procedure may be performed. Even when services are provided at a health facility in India, such as a PHC or post-partum center, the event is labelled a camp when either personnel (e.g. surgeons, anaesthesiologists) or medicines and equipment are externally supplied. Sterilization camps may be convened at various levels of the health system—from subcenters to district hospitals--as well as at public institutions such as schools where a building is available. Acceptors are recruited by both health workers and development sector staff, although services at these camps are provided by Ministry of Health and Family Welfare staff, who remain responsible overall for their quality.
Studies indicate that standards of infrastructure, equipment and logistical support vary according to the level of health facility where the camp is convened. Higher level facilities—e.g., Post-partum Centres and Community Health Centres--tend to have more adequate infrastructure, equipment and trained staff than do primary health centers and subcenters. In the latter, the operating theatre is usually a converted room with little audio and visual privacy. Outreach camps held at non-health institutions appear to have the least adequate facilities: they are often dark and offer little privacy; they are frequently without clean running water in the room serving as the makeshift operating theatre; and sometimes without electricity. In such cases, water has to be carried in and lighting obtained from battery operated flashlights or from power derived from a vehicle battery. In these camps and in those held at the smaller health centers where the electricity supply is irregular, researchers observed that there were occasions when surgery was interrupted for several minutes while waiting for the power situation to be resolved. Irrespective of type of setting, a finding common to all the studies on sterilization camps is that toilet facilities are either non-existent, or where are available, are either not functioning and/or filthy, an important issue given the practice of administering an enema to patients prior to the surgery.
Prior to surgery, the screening of patients by a systematic physical examination is essential in order to rule out high-risk cases. This should include the measurement of blood pressure, and urine and hemoglobin tests, as well as a general physical exam that involves a pelvic exam. The findings on whether screening was actually performed vary from none in the Bihar study (Parveen, 1995), to cursory examination without a pelvic exam in the Gujarat study (Mavalankar and Sharma, 1999), to performed at all sites which the researchers observed in Madhya Pradesh, Uttar Pradesh, and Kerala (Lakshmi R. and Barge, 1999; Khan et al. 1999a; Ramanathan, et al. 1995). In Madhya Pradesh, clients were screened by a medical officer following registration, and there were those who were subsequently rejected for sterilization surgery-- rejected clients averaged two to three per camp. In Uttar Pradesh, it was observed that in some settings, criteria qualifying women for surgical sterilization were not stringently enforced: women who were contraindicated for sterilization as a result of anemia were not rejected since service providers did not want to lose a motivated client or case that would contribute toward meeting assigned targets. A separate national study by the Indian Council of Medical Research also found highly inadequate screening of clients for potential contraindications at sterilization camps, and a universal failure to defer potential clients when such contraindications existed (ICMR, 1991).
Commonly observed shortcomings included the re-use of and failure to adequately sterilize needles and syringes for pre-operative medication, and the re-use of the same enema instruments without cleaning or washing (ICMR, 1991; Lakshmi R. and Barge, 1999; Mavalankar and Sharma, 1999). Another common finding was that the stipulated time interval between the administration of the medication and the surgery is often not observed, and the full effects of the drugs are not felt, causing the patient much discomfort. In Bihar, it was observed that particularly at outreach camps, there was occasionally an insufficient supply of anaesthesia for the case load, so that the last few patients were administered a lower and inadequate dosage (Parveen, 1995).
At the majority of camps, standard theater equipment was either non-existent or in extremely poor condition, and use of improvised equipment was reported in more than one study (e.g. makeshift operating table). Operating theatre apparel (gowns, masks, slippers) are in short supply so that these are not changed between patients. The adequacy in supply of gloves varies by facility, but it is clear from these studies that providers are not scrupulous about changing gloves between patients. Gowns are rarely provided to patients: they go in for surgery with their petticoats pulled up around them in lieu of gowns and are thus exposed to male staff.
Almost all studies—with the possible exception of Kerala—found that the maintenance of aseptic conditions during surgery was extremely poor. Again this appeared to vary according to the level of the facility, with the most glaring deficiencies in aseptic conditions reported at outreach camps. The importance of maintaining sterile conditions where surgical procedures are involved does not appear to be recognized by many providers. At the same time, this is exacerbated by the inadequacy of logistical support and supplies, the often heavy case load at these camps, and the time pressure providers tend to be placed under in these circumstances. The specific failures noted in observing aseptic precautions in the operation theater included changing gloves, gowns and caps after each procedure, the failure of health providers and surgeons to scrub prior to each operation, the re-use of the same operating sheets throughout the entire day of the camp, inadequate disinfection of surgical instruments between procedures, and the use of unsterilized sutures and dressings for the incision.
An Indian Council of Medical Research study found that 77 percent of operation theaters in its sample were not properly maintained or equipped. Moreover, in 40 percent of the sterilization procedures observed, surgical equipment and instruments were sterilized improperly or not at all (ICMR, 1991). Other studies have also documented major lapses in both hygienic conditions and proper sterilization of surgical equipment and supplies (Pettigrew, 1984; Lakshmi R. and Barge, 1999; Mavalankar and Sharma, 1999). Even though sterilization procedures and operation theater conditions appear to be somewhat better in the South Indian states, the study in Kerala still observed an absence of toilet facilities and the failure among staff to change surgical linens and gloves after each patient (Ramanathan et al. 1995).
At sterilization camps, the operating theatre commonly resembles an assembly line with two patients laid out on separate tables at the same time so that the surgeon can turn immediately from one to another as soon as he has completed the procedure. With a laparoscopy, the time taken from making the incision to the completed ligation is only two to three minutes, and the medical officer may suture and dress the incision so the surgeon can move on to the next patient. The study from Kerala, for example, found that the surgeon completed 48 sterilization procedures in just over two hours (Ramanathan, et al. 1995); other studies have reported equally high or even higher numbers of procedures performed (Pettigrew, 1984; Mehta, 1989; Bhatia, 1999).
Following the operation, patients are transferred to the recovery room and placed on beds, mattresses or rugs on the ground for relatives to look after. As stretchers are not available in most facilities, the inert and unconscious women patients are carried out in the arms of a male attendant to the recovery area. At most facilities where camps are held, the recovery area is not private. Medically, there is no monitoring of patients after surgery to ensure that they will not go into shock, and the surgeons depart as soon as the last operation has been performed, with only a cursory check of the patients, if at all. Although it is recommended that patients be discharged only four to five hours after surgery, all the studies noted that they were discharged much sooner, in some places even before they were fully conscious. Before discharge, patients are given a packet of medication containing analgesics and vitamins and/or iron supplements, and in some places antibiotics. They are given minimal verbal instructions about how to take care of themselves (e.g. not to get the wound wet) but they do not receive written advice. Patients and their family attendants are usually transported to their homes in a government-provided vehicle.
A number of studies from India have reported high levels of stated post-operative complications and extreme pain following acceptance of long-term contraception. A 1982-83 Indian Council of Medical Research study in Gujarat, Rajasthan, and Uttar Pradesh found that between 29 to 46 percent of IUD users and 12 to 23 percent of sterilization acceptors experienced method-related complications (ICMR, 1986). The 1988-89 Third All India Family Planning Survey found that 12 percent of temporary method acceptors, but as high as 36 percent of sterilization acceptors, reported problems after accepting the method or undergoing the procedure (ORG, 1990), the latter figure not dissimilar to that reported by Murthy (1999) for sterilization acceptors in Maharashtra. In the study in Uttar Pradesh, 47 percent of tubectomy acceptors and 30 percent of IUD acceptors reported post-acceptance complications (Khan et al. 1999a). An earlier study found that 23 percent and 13 percent of sterilized couples in Uttar Pradesh and Gujarat, respectively, reported complications after acceptance (Khan and Gupta, 1988). Similarly, a separate study by Visaria (1999) in Gujarat found that 26 percent of sterilization acceptors reported post-acceptance complications, and that 68 percent of these women continued to experience these problems. Findings from qualitative studies also attest to the fact that post-acceptance complications appear to be common and to have serious implications for women’s lives and well-being (Barge and Lakshmi R., 1999; Pettigrew, 1984). Whether these problems are actual or perceived, when combined with very low levels of follow-up, they have serious implications for how the program is viewed by its clientele.
One potential consequence of poor standards of technical quality of family planning care may be associated gynaecological morbidity, which has been found to be very high in a number of community-based studies in India (Koenig, et al. 1998). While it is very likely that much of this morbidity preceded rather than resulted from contraceptive usage (Char, 1999), there is at least some evidence of a possible association between contraceptive use status and increased gynaecological morbidity. A community-based study from South India, for example, reported an association between tubectomy and increased risks of vaginitis and painful menstruation (Bhatia, et al. 1997). A separate study of pelvic inflammatory disease in Bombay has hypothesized that invasive methods of fertility control (e.g., the IUD, sterilization, and abortion), rather than sexually transmitted diseases, may play a major causative role in such morbidity (Brabin, et al., 1998; Gogate et al. 1998). Moreover, the apparently not uncommon practice of inserting IUDs or providing sterilization in spite of a pre-existing reproductive tract infection, given the pressures upon workers to achieve family planning targets (Khan, et al. 1999a), may lead women to attribute a causal role to the family planning method with respect to their reproductive health problems.
Client Follow Up
Given high rates of method-related complications as well as associated reproductive morbidity, client follow-up represents an important component of high quality services. Most studies of this issue in India point to the absence of follow-up as an acknowledged failing of the government program. In the most recent National Family Health Survey, only 15 percent of IUD and pill users, and 30 percent of sterilization acceptors, reported having received home follow-up visits by health workers post-acceptance, with rates not markedly higher in the South and Western Indian states (IIPS, 1995). The need for such care, however, was indicated by the very high proportion of acceptors, especially of the IUD, who sought medical consultation following adoption of the method. Similarly, a study of clients in Kerala found that only seven out of the 22 cases observed at the PHC level were given a fixed time to revisit the PHC (Ramanathan, 1995). In the focus group study in Gujarat, a primary source of dissatisfaction among participants was the absence of follow-up and post-operative care, with allegations of frequent charging by workers for providing such services (Sharif and Visaria, 1991). Other qualitative studies also highlight the failure of the Indian program to provide adequate and effective follow-up care following sterilization (Pettigrew, 1984; Dharmalingam, 1995), an issue which assumes added importance in light of the previously noted high rates of reported post-operative complications and pain.
A 1982-83 Indian Council of Medical Research three-state study found that among women experiencing problems after undergoing sterilization, only a minority (30 to 43 percent) reported follow-up visits by government staff to assist in solving problems (ICMR, 1986), levels not dissimilar from a separate study in Uttar Pradesh and Gujarat (Khan and Gupta, 1988). Similar low levels of follow-up-- ranging from 11 percent in Bihar to a high of 34 percent in Kerala—were found in a subsequent study by the Council (ICMR, 1988). In a recent study in Uttar Pradesh, only about one-quarter of long-term method acceptors with complications reported that they received any help from a health worker in addressing these problems (Khan et al. 1999a).
Although proper follow-up is an essential component of quality care, a consistent finding has been that while ANMs consider it important to follow-up clients who have been sterilized, few follow-up IUD acceptors or consider it necessary to visit acceptors of oral contraceptives (Khan et al., 1999b). Workers visit sterilization clients to remove stitches or to refer them for further follow-up at the PHCs with a physician. Workers are also conscientious about following-up sterilization cases since it is a community expectation that the health worker visits a client after surgery. Sterilization acceptors in Gujarat complained that ANMs were not sympathetic to their complaints. ANMs, on their part, felt these complaints were related less to the surgery and more to the generally poor health condition of these women (Visaria, 1999). Follow-up of IUD acceptors is not a common practice among ANMs; few consider it necessary and instead instruct IUD acceptors to contact them if they encounter any problems. ANMs rarely follow-up acceptors of oral contraceptives, and assume that users should not have problems, and that if they do, they will either discontinue the method or contact a health worker.
A common finding has been that ANMs do not maintain their registers adequately to follow-up family planning acceptors, and lack a clear idea as to how many of their clients are still using a specific method, how many have discontinued and for what reasons. A national Indian Council of Medical Research study, for example, found that for 36 percent of sterilization acceptors and 43 percent of IUD acceptors, no record of the client was maintained at the Primary Health Center, a prerequisite for subsequent follow-up visits (ICMR, 1991). In the four-state study by Verma and Roy (1999), no more than 50 percent of ANMs in West Bengal and Karnataka— and even lower percentages in the other two states—maintained detailed registers which would allow the tracking of clients.
The ANM is required to visit each patient at home following sterilization surgery, often 2, 4 and 7 days after, to change the dressings, administer antibiotics if required, and after 7 days to remove the stitches. Studies have found that ANMs are generally conscientious about following-up sterilization acceptors and that their visits are regular (Townsend et al., 1999; Lakshmi R. and Barge, 1999). However, one study noted that ANMs are given no set protocol for these follow-up visits and thus have no guidelines with which to ascertain problems (Mavalankar and Sharma, 1999).
Appropriate Constellation of Services
The appropriate constellation of services includes the availability of both doctors and medicines, the timing of clinic hours and the waiting time to be seen by medical or paramedical staff, the location of services, and the adequacy of facilities. The presence of medical personnel, particularly of a doctor, is central to the availability of services. The findings of these studies show that doctors are not always present: studies in Kerala and Maharashtra found that only half of the primary health centers they studied had a physician in post (Ramanathan, 1995; Murthy, 1999). An in-depth study of a primary health center in Madhya Pradesh found that generally only one out of four doctors assigned to the PHC were actually on duty, and that on average this physician spent only two out of a mandated five hours providing services (Singh and Kumar, 1998). As the four-state study underscores, there also exists great variation by state. In Bihar, only 30 percent of respondents reported that a doctor is always available in a government facility when needed, compared to roughly two-thirds in Karnataka and Tamil Nadu (Roy and Verma, 1999). In Maharashtra, 64 percent of respondents stated that a doctor was always available in a government clinic when needed; this contrasts with 87 percent for physicians in the private sector (Murthy, 1999). Even in the South , Ravindran’s study in Tamil Nadu observed that doctors were generally not available for the duration of the clinic, but for two to three hours a day, after which they carried out private practice in the afternoons (Ravindran, 1999).
Client access to services is also heavily influenced by clinic timings--whether these hours are convenient to clients and whether these stated working hours are actually observed by staff. In these studies, the majority of respondents stated that the timing of the government health facility which they use was convenient (Roy and Verma, 1999; Barge and Lakshmi R., 1999). More detailed studies revealed lower levels of satisfaction with the timing of government health facilities. For example, a focus group study in Uttar Pradesh found a central reason behind dissatisfaction with government health services was that they coincided with work hours (Levine, et al. 1992). Whether the facility was actually open during the stated hours was a different matter. At one-half of the health facilities in the Tamil Nadu study, clinic timings were unpredictable (Ravindran, 1999). The reliability of opening hours appears to diminish as one moves to a lower-level facility (i.e., from primary health centers to subcenters) which tends to be poorer in infrastructure, equipment and supplies, staffed by a provider who is less likely to be resident, and less likely to be motivated since the services she can offer are limited given the inadequate resources.
Waiting time to consult a physician also varied considerably. In Bihar, although only 30 percent respondents reported that a physician was readily available at a government clinic, 50 percent reported a waiting time of under half an hour for consulting a doctor there (Roy and Verma, 1999). Exit interviews with women patients at government clinics in the four-state study found that fewer than 10 percent of women were forced to a wait for an hour or more to be seen by a doctor; the exception was West Bengal, where 21 percent of respondents reported such a lengthy wait (Roy and Verma, 1999).
A physician or a health facility’s perceived efficacy is governed by a successful response to a patient’s health care requirements and this, in turn, depends upon the provision of the appropriate medication. The four-state study by Roy and Verma (1999) indicates extensive variation between North and South Indian states in the availability of medicines at government clinics. Whereas 72 percent of women interviewed in Tamil Nadu believed that adequate medicines were always available at government clinics, only 50 percent of women interviewed in Karnataka expressed this view. Even in Tamil Nadu, however, Ravindran (1999) found that the acute shortages of drugs and supplies made it necessary for clients to purchase other than common curative medicines for pains and fevers, colds, coughs and diarrhoea from a pharmacy. Women being admitted to government facilities for institutional deliveries had to bring the home birthing kits distributed by ANMs given the shortage of supplies, and had to purchase not only blood and saline but gauze and cotton wool. Shortages of medicine were much more acute in the North Indian states, where Roy and Verma found that only 23 and 14 percent of respondents in Bihar and West Bengal, respectively, considered medicines to be regularly available.
The absence of appropriate public sector service facilities extends to other reproductive health services as well. A 1987-89 Indian Council of Medical Research study found that only 50 out of 200 Primary Health Centers sampled had adequate facilities for performing abortion, despite this procedure having been legal since the early 1970s (ICMR, 1991). Similar findings were reported in a more recent four-state situational analysis of abortion facilities in India (Barge et al. 1998).
A client’s reported satisfaction with services is regarded by many observers as an important indicator of the quality of care received. A seeming paradox from Indian studies are the generally very high reported rates of client satisfaction with existing services, despite the numerous and serious deficiencies in quality of care highlighted within this review. In the most recent National Family Health Survey (1992-93), for example, 53 percent of sterilization acceptors rated the care they had received during the procedure as very good or excellent, and 39 percent felt it was ‘alright’; only seven percent of clients rated their care as ‘not so good’ or ‘very bad’. Comparable ratings were evident for sterilization acceptors’ assessment of the follow-up care they received (IIPS, 1995). Similarly, 97 percent of sterilization acceptors in Gujarat, and 71 percent of acceptors in Uttar Pradesh, had recommended the method to other prospective users (Khan and Gupta, 1988). An earlier study by the Indian Council of Medical Research also found that a very high proportion of visited women —ranging from 60 percent in Kerala to 79 percent in Bihar to 94 percent in Gujarat—held positive views about workers in the family planning program ICMR, 1988). Qualitative findings in Gujarat and Kerala also suggest low levels of dissatisfaction by clients in the face of indifferent quality of services (Ramanathan, 1995; Visaria and Visaria 1992).
Not all studies report such high levels of client satisfaction. In an earlier study in Uttar Pradesh, 43 percent of respondents visited by a fieldworker indicated that they would welcome a revisit by the worker ‘somewhat’ or ‘not at all’, with one-fourth falling into the latter category (Khan et al. 1994). In a more recent study by Khan, et al. (1999a) in Uttar Pradesh, only 45 percent of sterilization acceptors and 38 percent of IUD acceptors indicated a willingness to recommend these methods to others.
There are several plausible explanations for this apparent inconsistency between service quality and reported client satisfaction. First, clients are with few exceptions less well-educated than service providers, especially doctors, and are therefore unwilling or unable to question their treatment. Second, clients may have relatively few options in terms of health and family planning services, especially in rural areas, and are therefore more likely to accept existing standards as reasonable. Third, and perhaps most importantly, clients may have very low expectations in terms of services provided by the public sector in general, particularly since they are ostensibly provided free of cost, and may thus be easily satisfied with low quality of services. Murthy (1999), for example, found that clients were not able to adequately discriminate between high and low quality care. Clearly, more work is warranted on the factors which influence client perceptions of service quality, and the relationship of quality to client satisfaction, in India and elsewhere.
Our review has provided a realistic albeit bleak picture of the Indian family planning program at the ground level— with respect to service access and availability, quality, and actual implementation. The studies considered collectively highlight the major shortcomings in quality which confront the Indian program—in the limited contraceptive choices afforded to couples who wish to space or limit childbearing, in the significant gaps in counseling and information provided to clients, in the nature and tenor of client-provider relations, and in the extent of follow-up and continuity of care. Of equal concern are the apparently widespread gaps in technical quality of care and breaches in infection control, which potentially place clients at risk of a range of infectious diseases, a growing concern within India. Our review also underscores the fundamental but frequently overlooked point that access to and availability of services-- often as a precursor to concerns about quality of care-- remain significant issues for much of India’s rural population.
Reference is often made to the marked geographical diversity which exists within India on a number of social and cultural factors which influence the demand for fertility limitation (Dyson and Moore, 1983; Satia and Jejeebhoy, 1991). The studies reviewed here also highlight the significant geographical variability which exists in many aspects of service delivery—the lower levels of worker-client contact, infrastructure support, and rapport and affinity between clients and service providers consistently noted in the large North Indian states. Equally striking, however, is the similarity in quality of care concerns across regions-- most notably with respect to method choice, information provided to clients, clinical standards and procedures, and follow-up/continuity of care — suggesting that many quality of care concerns represent generic problems throughout the Indian family planning program.
The evidence we have reviewed illuminates four broad sets of interacting and reinforcing factors which constrain the accessibility and quality of family planning services within India’s public sector. Despite public sector commitment to family planning, resource limitations and underdevelopment remain serious barriers to effective service delivery at virtually all levels of the delivery system—as manifested in significant underfunding of basic program infrastructure especially at the peripheral level, an absence of educational, housing, and transportation facilities for program staff, the inability to safeguard the personal security of outreach staff, and chronic shortages of most basic medicines, equipment, and supplies. Weak program management capacity presents a second serious set of constraints-- as reflected in inadequate staff training, weak supervisory support for workers, frequent failure to adhere to acceptable clinical standards, limited accountability among program personnel, and widespread corruption.
It is also apparent that the basic philosophy and orientation of the Indian family planning program-- as reflected in the pervasive system of family planning targets for workers as basic performance indicators, the emphasis of quantity over quality, and the corresponding low priority attached to the needs of individual clients— must also be accorded a central explanatory role in the substandard levels of care within the program. A final key but often overlooked barrier to improved quality of care relates to client expectations and empowerment. It is striking that significant numbers of clients express satisfaction with existing family planning services, despite what could frequently be regarded as substandard levels of care by Western standards. The absence of client demand and expectations for at least some minimum standard of service—over and above simply receipt of the service sought—represents a crucial missing impetus for reform and greater concern within the public sector for higher quality, client-centered services.
The findings from the studies considered strongly suggest that an absence of concern for client needs and overall quality of care has played an important explanatory role in the disappointing performance of the Indian family planning program to date. The potential ‘costs’ of poor quality services can also be readily discerned from many of the studies considered— as reflected in a poor image and general distrust of the public sector system, a marked preference for the private sector among clients, and weak commitment and low esprit de Corp among government family planning staff. Although empirical evidence on the impact of quality of care remains extremely slender within India (Prabhavathi and Sheshadri, 1988-89; Mari Bhat and Halli, 1998; Patel et al. 1999) and elsewhere, it is very likely that poor quality of care has contributed to high levels of foregone, delayed, or discontinued use of contraception and consequent unwanted pregnancy, among many current or potential clients.
Considerable international attention has focused upon recent efforts by the Government of Indian to improve the quality and range of services provided through the Family Welfare Program. Following the 1994 International Conference on Population and Development, the Indian Government has made a serious effort to reorient existing family planning services toward a greater concern for quality of care and clients’ broader reproductive health needs. Prominent steps include a 1996 policy decision to withdraw the nationwide system of contraceptive method targets, which have largely defined the family planning program for the last three decades, and the launching in 1997 of the new Reproductive and Child Health Program, replacing the much narrower programs on maternal and child health and family planning. These policies aimed at fundamentally reorienting the family planning program toward a client-centered approach represent positive and long overdue changes, and are likely to have a positive impact in those areas of India where the program is fully operational. At the same time, it is important to recognize that by themselves, these policy changes fail to redress systemic and complex problems of implementation which continue to plague much of the Indian program. As our review has highlighted, many of these problems defy simple or ready intervention. As such, reorienting Indian family planning services toward a higher quality, more client-centered approach represents a formidable task. Progress toward this end is likely to prove painstakingly slow, and will require an extended timeframe measured more in decades than in years.