RESEARCH
Evidence-based reproductive health care in Cameroon: population-based study of awareness, use and barriers
Soins de santé génésique fondés sur l'expérience au Cameroun : études en population de la connaissance et de la mise en œ uvre de ces soins, ainsi que des obstacles rencontrés dans cette mise en œ uvre
Atención de salud reproductiva basada en la evidencia en el Camerún: estudio poblacional sobre los conocimientos, la aplicación y las barreras
Alan TN TitaI,1; Beatrice J SelwynII; D Kim WallerII; Asha S KapadiaII; Sylvestre DongmoIII
IMFM Division, Obstetrics and Gynecology, University of Alabama at Birmingham, 618 20th St South (OHB 458), Birmingham, AL 35233, USA. At the time of the study this author was affiliated with Baylor College of Medicine, Houston, TX, and the University of Texas Health Science Center at Houston, Houston, TX, USA
IISchool of Public Health, University of Texas Health Science Center at Houston, Houston, TX, USA
IIIBafut Health District, North West Provincial Delegation of Health, Bamenda, Cameroon
ABSTRACT
OBJECTIVE: To estimate the prevalence of awareness and use of evidence-based reproductive health interventions and to describe the barriers associated with the use of evidence-based interventions among health providers in north-west Cameroon.
METHODS: In February 2004, a population-based descriptive study of the awareness and use of 13 evidence-based interventions targeted health workers providing reproductive health care. Their awareness and use of a composite of four vital interventions was also evaluated. These were peripartum use of antiretrovirals to prevent transmission of HIV, antenatal corticosteroid administration, magnesium sulfate prophylaxis and active management of placental delivery with uterotonics. In-depth interviews with key informants were conducted as part of a qualitative substudy to discover the barriers to the use of evidence-based interventions.
FINDINGS: Overall, 91.4% (328/359) of reproductive health workers were surveyed. Their awareness of evidence-based interventions varied from 29% for the use of antenatal corticosteroids to 97% for the use of iron and folic acid supplementation during pregnancy. Their use of these interventions ranged from 10.2% for antenatal corticosteroids to 94.8% for iron and folic acid supplementation. Only 50/322 (15.5%; 95% confidence interval (CI) = 11.820.0) of health workers were aware of all four vital interventions, and only 12/312 (3.8%; 95% CI = 2.0 6.6) reported using all of them regularly.
A total of 26 key informants participated in the qualitative substudy. A deficiency in the education and training of health workers, especially a lack of continuing education, was commonly identified as the most important barrier to their awareness of evidence-based practices. A lack of awareness and a lack of supplies and materials were the main barriers to practice.
CONCLUSION: The awareness and practice of important evidence-based reproductive health interventions were less than optimal. To improve maternal and perinatal outcomes both remedial programmes to enhance awareness, including continuing education for health workers, and the provision of necessary supplies are needed.
Keywords: Reproductive health services; Evidence-based medicine; Health personnel/education; Health knowledge, attitudes, practice; Pregnancy complications/drug therapy; Adrenal cortex hormones/therapeutic use; Anti-retroviral agents/therapeutic use; Magnesium sulfate/therapeutic use; Tocolytic agents/therapeutic use; Population surveillance; Cameroon (source: MeSH, NLM).
RÉSUMÉ
OBJECTIF: Estimer la fréquence de la connaissance et de l'utilisation d'interventions de santé génésique reposant sur l'expérience et décrire les obstacles s'opposant à la mise en oeuvre de ces interventions par les soignants au nord-ouest du Cameroun.
MÉTHODES: En février 2004, une étude descriptive en population de la connaissance et de l'utilisation de 13 interventions reposant sur l'expérience a été menée chez des agents de santé dispensant des soins de santé génésique. On a également évalué la connaissance et l'utilisation d'un ensemble de quatre interventions vitales à savoir, l'utilisation pendant le péripartum d'antirétroviraux pour prévenir la transmission du VIH, l'administration prénatale de corticostéroïdes, la prophylaxie par le sulfate de magnésium et la prise en charge active de la délivrance placentaire au moyen d'utérotoniques. Des entretiens approfondis avec des informateurs clés se sont déroulés dans le cadre d'une étude subsidiaire qualitative, visant à déterminer les obstacles à l'utilisation d'interventions reposant sur l'expérience.
RÉSULTATS: Au total, 91,4 % (328/359) des soignants en santé génésique ont participé à l'enquête. Leur connaissance des interventions reposant sur l'expérience se situait entre 29 % pour l'utilisation prénatale de corticostéroïdes et 97 % pour l'utilisation de la supplémentation en fer et en acide folique au cours de la grossesse. La mise en œ uvre de ces interventions se situait entre 10,2 % pour l'administration prénatale de corticostéroïdes et 94,8 % pour la supplémentation en fer et en acide folique. Seuls 50 agents de santé sur 322 (15,5 % ; intervalle de confiance [IC à 95 % = 11,8-20,0) connaissaient les quatre interventions vitales et seuls 12/312 (3,8 % ; IC 95 % = 2,0-6,6) ont indiqué qu'ils les utilisaient tous régulièrement.
Au total 26 informateurs clés ont participé à l'étude subsidiaire qualitative. Des lacunes au niveau de l'éducation et de la formation des agents de santé, notamment l'absence d'une formation continue, ont été fréquemment identifiées comme l'obstacle le plus important à la connaissance des pratiques reposant sur l'expérience. Les connaissances insuffisantes et le manque de fournitures et de matériel étaient les principaux obstacles à leur mise en œ uvre.
CONCLUSION: La connaissance et la mise en œ uvre d'interventions importantes en santé génésique et reposant sur l'expérience n'étaient pas optimales. Pour améliorer les résultats obtenus au niveau maternel et périnatal, il faut à la fois offrir des programmes pour renforcer les connaissances des agents de santé, notamment une formation continue, et apporter les fournitures nécessaires.
Mots clés: Services de médecine de la réproduction; Personnel sanitaire/enseignement; Médecine factuelle; Connaissance, attitude, pratique; Grossesse compliquée/chimiothérapie; Hormones corticosurrénaliennes/usage thérapeutique; Agents antirétroviraux/usage thérapeutique; Sulfate magnésium/usage thérapeutique; Agents tocolytiques/usage thérapeutique; Surveillance population; Cameroun (source: MeSH, INSERM).
RESUMEN
OBJETIVO: Estimar la prevalencia de los conocimientos y la aplicación de las intervenciones de salud reproductiva basadas en la evidencia, y describir las barreras que se oponen al uso de esas intervenciones entre los proveedores de asistencia sanitaria en el noroeste del Camerún.
MÉTODOS: En febrero de 2004 se realizó un estudio descriptivo poblacional sobre el conocimiento y uso de 13 intervenciones basadas en la evidencia, centrado en personal sanitario que dispensaba atención de salud reproductiva. Se evaluaron también sus conocimientos y aplicación de un conjunto de cuatro intervenciones fundamentales: el uso de antirretrovirales durante el parto para prevenir la transmisión del VIH, la administración prenatal de corticosteroides, la profilaxis con sulfato de magnesio y el fomento activo de la expulsión de la placenta con uterotónicos. Además, se llevaron a cabo entrevistas detalladas con informantes clave como parte de un subestudio cualitativo destinado a identificar las barreras a la aplicación de esas intervenciones basadas en la evidencia.
RESULTADOS: Se estudió en total al 91,4% (328/359) de los trabajadores de salud reproductiva. Su conocimiento de las intervenciones basadas en la evidencia variaba entre el 29% para la utilización prenatal de corticosteroides y el 97% para la administración de suplementos de hierro y ácido fólico durante el embarazo. Su utilización de esas intervenciones variaba desde el 10,2% en el caso de los corticosteroides prenatales hasta el 94,8% en el caso de los suplementos de hierro y ácido fólico. Sólo 50 de 322 (15,5%; intervalo de confianza [IC] del 95% = 11,820,0) trabajadores sanitarios conocían las cuatro intervenciones esenciales analizadas, y sólo 12 de 312 (3,8%; IC95% = 2,06,6) señalaron que aplicaban todas ellas regularmente.
Un total de 26 informantes clave participaron en el subestudio cualitativo. La barrera más importante en cuanto al conocimiento de las prácticas basadas en la evidencia resultó ser una deficiente formación teórica y práctica del personal sanitario, especialmente la falta de enseñanza continua. La falta de conocimientos y la falta de suministros y material eran las barreras principales a la aplicación de esas prácticas.
CONCLUSIÓN: El conocimiento y la aplicación de importantes intervenciones de salud reproductiva basadas en la evidencia eran subóptimos. Para mejorar los resultados maternos y perinatales se necesitan tanto programas correctivos que aumenten los conocimientos, en particular enseñanza continua para los agentes de salud, como suministros suficientes.
Palabras clave: Servicios de salud reproductiva; Personal de salud/educación; Medicina basada en evidencia; Conocimientos, actitudes y práctica sanitarias; Complicaciones del embarazo/quimioterapia; Corticosteroides/uso terapéutico; Agentes antirretrovirales/uso terapéutico; Sulfato de magnesio/uso terapéutico; Agentes tocolíticos/uso terapéutico; Vigilancia de la población; Camerún (fuente: DeCS, BIREME).
Introduction
More than 500 000 women die annually from pregnancy-related complications and millions become ill from these complications (1). Altogether, 99% of these deaths occur in the developing world, especially Africa, where the average risk of pregnancy-related death is 1 in 13, compared with 1 in 4085 in developed countries (1). Also, 98% of neonatal deaths occur in developing countries: 42 neonates die per 1000 live births in Africa compared with at most 10 neonates per 1000 live births in developed countries (2). More than 70% of maternal deaths are caused by haemorrhage, infection, unsafe abortion, eclampsia and obstructed labour (3). Strategies proposed to improve maternal health include having skilled attendants present at delivery, having high quality reproductive health services available, allowing women to prevent unwanted pregnancies and giving them access to safe abortions (4). It is vital to use an evidence-based approach to maternal and neonatal health in areas where resources are constrained (5). Access to reliable evidence-based information is tenuous in developing countries (6), even though it is perceived as providing the most cost-effective and achievable strategy for sustaining improvements in health care (7). The WHO Reproductive Health Library (8) provides evidence-based information that is relevant to developing countries and was developed specifically to bridge the knowledge gap. However, information about whether health workers are aware of or use evidence-based interventions is mostly anecdotal.
We undertook a study in Cameroon to establish baseline information prior to intervening to improve evidence-based care. We focused on the English-speaking North-West Province where there are 1.8 million inhabitants and 32 000 deliveries annually. The coverage of antenatal care is 80%; more than 60% of deliveries are attended by a trained professional (9); and the maternal mortality ratio is estimated at 730 deaths per 100 000 live births (10).
Our objectives were, first, to estimate the prevalence of the awareness and use of certain evidence-based reproductive health interventions among health-care providers in this part of Cameroon. We also assessed their awareness and use of a composite intervention comprising four interventions we considered to be vital. Our second objective was to identify barriers to the awareness and use of evidence-based interventions.
Methods
Design and participants
A population-based survey of all reproductive health workers in the North-West Province of Cameroon was conducted in February 2004. Follow-up in-depth interviews were conducted with a subset of health workers in order to discover barriers to the uptake of evidence-based health interventions (11, 12).
The institutional review boards of Baylor College of Medicine, Houston, TX, and the University of Texas, Houston, and the Provincial Delegate for Health in Cameroon approved the study's protocol. All participants gave informed consent.
Evidence-based reproductive interventions
Version 5 of the WHO Reproductive Health Library, which is an extract of the Cochrane Library, contained 70 systematic reviews of reproductive health interventions, including commentaries, discussions and recommendations relevant to developing countries. We selected 11 interventions that had evidence of effectiveness, one with evidence of harm (routine episiotomy) and one that was of uncertain practical benefit in resource-poor settings (routine Caesarean for breech birth at term). These 13 interventions and their key effects are presented in Table 1 (1327).
We considered a composite of four interventions to be vital: management of placental delivery with uterotonics, the administration of magnesium sulfate to prevent convulsions in eclampsia, the antenatal use of corticosteroids for impending preterm delivery, and the peripartum administration of anti-retroviral drugs to prevent the transmission of HIV. The first two interventions address major causes of maternal mortality, and the latter two address major causes of perinatal mortality and morbidity. We assessed whether health workers used all of these interventions routinely and whether they were aware of all four of them.
Instruments and procedures
A three-part, 40-item pretested self-administered questionnaire, which used Dillman's validity enhancing strategies (28), was administered. Health units providing reproductive health care (prenatal, intrapartum and postpartum services) in all 14 health districts were targeted. These included hospitals and health centres run by the government, as well as missionary (nongovernmental) hospitals and health centres and private hospitals and health centres. The target population was all physicians, health workers in charge of health units or health centre maternity units, and all health workers at maternity units at district and provincial hospitals. Those health workers included were actively providing reproductive health care at the time of the study.
Most participants at health centres and hospitals were surveyed together during scheduled meetings at district hospitals. Absentees were subsequently individually enrolled. Respondents were encouraged to write comments on questionnaires to clarify their answers; questionnaires were screened for completion by the respondent prior to placement at a neutral spot to assure anonymity. Surveys were followed by an educational debriefing session during which WHO recommendations were discussed. Debriefing also served as a way to monitor the validity of the questionnaires.
Only those who had taken part in the questionnaire survey and debriefing were eligible to be enrolled in the qualitative substudy. We selected a purposive sample of 26 workers in charge of health units or maternity sections, representing a mix of professional categories, types of facility (private or governmental and hospitals or health centres) and location within the province. All of those approached agreed to participate. The interviews were conducted by one of the authors (ATNT) in English, the official language in the province. An interview guide presenting key topics to be explored was used. Demographic information was collected as was information on barriers that might prevent health workers from knowing about or using evidence-based interventions. Interviews were tape-recorded and handwritten notes were also made; interviews were conducted in private and lasted 2040 minutes. A trained assistant performed verbatim transcription of the tapes. Tapes and transcripts were completely reviewed and compared with written notes.
Data management and analysis
A Microsoft Access 2000 database was used for survey management and quality control. If it was unequivocally clear from a written comment on the questionnaire that the wrong response had been selected, the correct option was coded and entered into the database. Responses were categorized into dichotomous outcomes corresponding to the presence or absence of the awareness or use of evidence-based interventions. Data were analysed using SPSS statistical software, version 11.0 (SPSS, Chicago, IL) and Stata, version 6.0 (Stata, College Station, TX).
Analysis of the qualitative data was done using open coding, data reduction, categorization, axial coding and selective coding. Methods of case analysis and cross-case analysis were applied to identify patterns and salient themes in the data (29).
Sample size
Preliminary information indicated there were about 350 health workers who met the inclusion criteria. This sample size was sufficient to use awareness of the composite vital interventions as the primary outcome with a validated formula (30): 385 participants would be needed to estimate a prevalence of 50% to a precision of ± 5%, and only 96 would be needed for a precision of ± 10%.
Enrolment in the qualitative substudy was stopped after 26 interviews because major themes were recurring frequently.
Findings
A total of 328 reproductive health workers in 188 of 204 (92.2%) eligible health units in North-West Cameroon were surveyed, including 54 of 63 (85.7%) physicians and 274 of 296 (92.6%) non-physicians. Units not represented in the survey were physically or administratively inaccessible or had workers who were absent or excluded because they had inadvertently participated in a debriefing session prior to being surveyed. Only one health worker did not return a survey. A total of 108 participants (32.9 %) wrote comments. At least one response was re-coded on 10 questionnaires (3.0%) based on unequivocal comments.
Descriptive characteristics
Characteristics of survey respondents are presented in Table 2 (web version only, available from www.who.int/bulletin). Reproductive health workers had a mean age of 40.9 years and a mean of 13.3 years of work experience. Few were obstetricians: most were government workers and only 25% had access to a library or to the Internet or email. Awareness of the Reproductive Health Library was low, and access to it was limited. On average, health workers read a journal or accessed a database only three times per year, and they attended fewer than two educational events over three years: most had done neither.
A total of 13 female and 13 male health workers in charge participated in the qualitative substudy: 9 nurses, 10 midwives and 7 physicians (2 obstetriciangynaecologists). Sixteen of the participants were government employees; three worked in the private-for-profit sector; and 7 worked at facilities run by missionaries or nongovernmental organizations.
Awareness and use of evidence-based interventions
In Table 3, the prevalence of the awareness and use of evidence-based interventions are presented. The levels of awareness of 9 interventions and the use of 11 were below 80%; levels were below 50% for up to 6 and 8 interventions, respectively. Awareness of the vital interventions was extremely low: only 15.5% (50/322) of participants knew about all four of them. Awareness was lowest for the use of antenatal corticosteroids: only 29.0% (95/328) knew about this intervention. The use of magnesium sulfate to treat eclampsia was another vital intervention that was not well known (48.9%; 159/325). Awareness of emergency contraception was poor (34.0%; 111/326) as was knowledge of periconceptional folate (39.0%; 127/326), external cephalic version (39.3%; 129/328) and the provision of social support during labour. The best-known intervention was the use of iron and folic acid supplementation during pregnancy (97.0%; 318/328).
The use of the four vital interventions was extremely low: only 12/312 (3.8%; 95% confidence interval (CI) = 26.6%) of participants used them. The intervention that was least likely to be used was antenatal corticosteroids (33/323; 10.2%). Magnesium sulfate was not used often (79/319; 24.8%) nor was peripartum antiretroviral treatment (130/325; 40%). Uterotonics were likely to be used (231/323; 71.5%). Iron and folic acid supplementation was the intervention that was most likely to be used (307/324; 94.8%). For 12 of the 13 interventions, prevalence of awareness was higher than prevalence of use, with peripartum antiretroviral treatment showing the largest difference between awareness and use. Interestingly, the practice of avoiding routine episiotomy exceeded awareness of the evidence (86% avoided the use of it versus 72% who were aware of the evidence).
Perceived levels of awareness and use
Health workers' perceptions of the awareness and use of the 13 interventions by other health workers were elicited during interviews and are presented in Table 4. Their responses confirmed that there was both a lack of awareness of the evidence for and lack of use of antenatal corticosteroid therapy, magnesium sulfate for seizure prophylaxis and emergency contraception. They also confirmed that there was universal use of uterotonics and iron and folate supplementation. Also, the practice of avoiding routine episiotomy was perceived to be higher than the awareness of the evidence for avoiding it. Participants believed that almost no one used periconceptional folate supplementation to prevent neural tube defects.
Quotations highlighting the barriers to the awareness and use of evidence-based interventions are presented in Box 1 and discussed below.
Barriers to awareness
The overwhelming theme of the interviews was that the education and training of health workers was deficient.
Deficiencies in education and training were cited by 21 of the 26 health workers as a general barrier. They were also the most frequently cited barriers to the awareness of nine of the interventions, including the use of antenatal corticosteroids, emergency contraception, periconceptional folic acid and prophylaxis with magnesium sulfate. Four underlying barriers were identified, the first being the absence of continuing education (Box 1, quote 1). This was the most frequently cited general and specific barrier. Suggestions made about means to rectify this deficiency and improve the awareness of evidence-based interventions included the use of seminars, refresher courses, daily physician-led ward rounds and weekly lectures at the hospitals. The second quotation in Box 1 illustrates this common sentiment.
The second underlying barrier was a lack of adequate formal training. This barrier was attributed primarily to medical and nursing school curricula that failed to emphasize evidence-based care.
The third underlying barrier was a lack of access to educational resources. Participants said they had difficulty gaining access to health journals, newsletters, databases and books. This was seen to contribute to their lack of awareness. Only one public library (run by the British Council) existed in the province.
The fourth underlying barrier was a lack of a habit of self-learning. Some participants cited the lack of a fundamental culture that encouraged reading and updating knowledge as another underlying factor (Box 1, quote 3).
Participants cited a lack of exposure to evidence-based interventions as another barrier to the awareness of these interventions. This response was given to explain the lack of awareness of interventions, such as corticosteroids, magnesium sulfate prophylaxis, social support during pregnancy and emergency contraception. Altogether, 21 health workers attributed their lack of awareness of the use of caesarean section for breech births at term to the widespread view of the breech position as a normal presentation (Box 1, quote 4).
Some health workers felt that the advent of new evidence-based interventions led to a lack of awareness because new information was not disseminated in a timely fashion (Box 1, quote 3).
Other barriers that were cited less frequently included the disbelief or denial of the existence of certain diseases, such as HIV/AIDS; the implementation of government policies that fail to promote evidence-based interventions; and the existence of unqualified supervisors who are unable to educate their subordinates.
Barriers to use
The barriers to the use of evidence-based interventions were primarily a lack of awareness of these interventions and a lack of necessary materials and supplies.
Lack of awareness of interventions. The lack of awareness of evidence-based interventions was the barrier that was most frequently cited and was perceived to impede the use of 11 interventions. It was cited by all 26 health workers to explain the low use of periconceptional folic acid.
Lack of supplies. A lack of necessary supplies was felt by many health workers to impede their use of evidence-based medicine. For example, they cited a lack of magnesium sulfate, as well as a lack of testing materials and medications for HIV, and a shortage of operating rooms where there was adequate equipment to perform Caesarean sections.
Negative sociocultural beliefs. Negative sociocultural beliefs were believed to impede the practice of interventions such as the administration of peripartum antiretroviral treatment, the provision of companionship during labour and the routine use of Caesarean sections for breech births. Participants stated that patients refused to accept testing, treatment and other services associated with HIV because of stigma and taboos. Male companionship during labour and delivery was taboo in some cultures. Other companionship was often refused because of concerns about witchcraft, privacy and gossip. Refusal of Caesarean sections was frequently related to the belief that the surgery portends reproductive failure (Box 1, quote 9).
Financial barriers. Poverty at all levels (state, family and individual) was also described as a barrier to the use of evidence-based interventions. State poverty ultimately leads to a lack of necessary supplies. Patients' inability to afford treatments, including antibiotics (for Caesarean sections or asymptomatic bacteriuria), and the high cost of Caesarean sections were also cited.
Other barriers. The other barriers cited by participants encompassed a diverse range of issues. They are outlined below.
· The widespread use of alternative interventions was perceived to be a deterrent to evidence-based practice. The use of less efficacious antiseizure regimens prevented magnesium sulfate prophylaxis; the use of vaginal delivery for breech births prevented the use of Caesarean sections.
· Religion was also perceived to be a barrier. For example, the use of emergency contraception is prohibited at Catholic health units.
· A lack of promotion by the government was blamed for the non-use of interventions such as antenatal corticosteroids, magnesium sulfate and periconceptional folic acid.
· A shortage of skilled health staff also prevented the use of some interventions. For example, a shortage of physicians contributed to the lack of use of Caesarean sections and external cephalic version for breech births. The retirement of older midwives limited the use of some interventions (Box 1, quote 6). A lack of skills resulted in an avoidance of routine episiotomy (Box 1, quote 7).
· A lack of access to reproductive health units was perceived to prevent the administration of peripartum antiretroviral treatment, antenatal corticosteroids and the use of Caesarean section or external cephalic version for breech births.
· Barriers among staff and in health units were also identified. Misdiagnosis, misconceptions (Box 1, quote 9), the insecurities of health workers, paternalistic perceptions (quote 8), and obligations to maintain privacy prevented the practise of some interventions, including external cephalic version and the use of a companion during labour.
· A lack of demand was also cited, specifically with regard to emergency contraception.
Discussion
Overall, the awareness and use of evidence-based interventions in reproductive health care was poor, although not uniformly so. Use consistently lagged behind awareness, except in the case of routine episiotomy for which practice paradoxically exceeded awareness of the evidence. Only 15% of participants were aware of all four of the key interventions likely to yield the largest benefits in preventing maternal and perinatal mortality and morbidity, and only 34% of participants used all four interventions. Among interrelated barriers, deficiencies in education and training, especially a lack of continuing education, were the principal barriers to awareness. A lack of awareness of evidence-based interventions and a lack of requisite supplies and materials were the leading barriers to use.
Our findings are consistent with those of a hospital-based study in China that concluded obstetric practice did not follow the best available evidence (31). Variation in the awareness and use of individual interventions is reflected by the 82% rate of use of routine episiotomy in the study in China compared with the 85.5% avoidance rate we found in Cameroon.
We gained insight into the reasons for the wide variation in the awareness and use of interventions. Many variables interact to bring about the variations; these include intervention-specific differences in education and training, the availability of supplies, government policy, the public health importance of related diseases, resource mobilization and duration of availability of evidence supporting interventions.
The fact that many health workers were aware of the use of peripartum antiretroviral drugs (such as nevirapine) to prevent mother-to-child transmission of HIV was likely to be the result of general government policies designed to sensitize the population to HIV/AIDS. No governmental programme to prevent transmission was active. However, the Baptist health service, through an internationally funded programme to prevent HIV transmission from mother to child, was considered to have played a leading part in enhancing the awareness and practise of this intervention. The programme provided ongoing training to government and private health workers as well as supplies (including nevirapine) at no cost to clients. Consequently, despite the 50% lag relative to awareness, the fact that 40% of participants reported using this intervention represented a tremendous improvement from 3% in 2001 (32). Eclampsia and neural tube defects were familiar to health workers as were low birth weight and prematurity, which contributed significantly to infant mortality (33). However, few health workers were aware of the interventions for these conditions, although they have been available for more than a decade. Some health-care workers attributed this poor awareness to the "recent" advent of the interventions. Obviously, educational deficiencies contribute too. Generally, a lack of use was primarily related to a lack of awareness. Thus, despite their widespread availability and postnatal use, corticosteroids were not given antenatally for prematurity. Magnesium sulfate was unavailable to the few physicians aware of its efficacy. The government had not promoted these interventions by including them in essential service packages unlike the interventions that were more likely to be used (iron and folic acid supplementation).
While the awareness of evidence-based interventions does not always translate into use, as indicated by the consistent lag in the prevalence of use relative to awareness, our results also show that higher levels of awareness are associated with higher levels of use. The paradox that higher levels of avoidance of routine episiotomy existed than did awareness of the evidence was explained by the lack of the requisite skills.
Limitations to our study include the fact that assessment of use was based on participants' own reports. We expected health workers to over-report practices. In one study in the United States over-reporting accounted for 50% of positive responses (34). Consequently, our results are conservative. The ability of qualitative methods to determine the relative importance of all barriers is limited (11). Barriers affecting patients and the non-medical community may not have been completely captured.
We assessed information bias as it related to two of the interventions. First, debriefing sessions revealed that 3/23 respondents mistook periconceptional folic acid for folic acid and iron supplementation. Adjusting for this mistake reduced the prevalence of awareness of periconceptional folic acid to at most 26% and of the use of it to 14%. Second, 14/108 respondents who wrote comments had probably over-reported the planned use of Caesarean sections for breech births. Adjustment reduced awareness to at most 50.7%. Nevertheless, the implication of a policy of routine caesarean section for breech birth is uncertain in such resource-poor settings where there is limited access to equipped operating rooms.
Our study revealed an opportunity for improving poor maternal and perinatal outcomes by addressing low and discrepant levels of awareness and use of effective reproductive health interventions (35). It is possible for levels of awareness and use to exceed 95%, as was seen in the case of iron and folic acid supplementation. Evidence-based medicine improves health outcomes, fosters cost effectiveness and increases patients' satisfaction (3641), despite some criticisms that it promotes orthodoxy and could be exploited to ration health care (36, 42).
There is an urgent need to design programmes to improve the awareness and use of evidence-based reproductive health care in countries like Cameroon. Such programmes would benefit from information regarding potential barriers to uptake (43) and should apply strategies proven to sustain continuing education (44). (44, 45). In complex scenarios, where there are multiple interacting factors, improved awareness of evidence-base interventions alone may be insufficient to change practice. Therefore, strategies to enhance their use must establish mechanisms to maintain the availability of the supplies and materials necessary to implement interventions. The key is to develop a comprehensive approach that extends the focus beyond health workers to the practice setting and wider environment and that is based on the specifics of evidence-based interventions; associated barriers and incentives to change must also be addressed (44). Such an approach, which applies multiple interventions, is more likely to achieve sustainable change (46). 
Acknowledgements
We thank Joe Leigh Simpson for advice during the planning and implementation of the study and critiqued the write-up of the study. We are also grateful to A Ndiforchu, Delegate of Health for the North West Province of Cameroon and his team of district health directors for their collaboration.
Funding: This study was funded primarily through a Johnson & Johnson dissertation grant in women's health from the Woodrow Wilson National Fellowship Foundation supplemented by a residency research project grant from the Arthur M Faris Fund at Baylor College of Medicine.
Competing interests: none declared.
References
1. UNICEF. WHO/UNICEF/UNFPA estimates for maternal mortality, 1995. New York (NY): UNICEF; 2003.
2. Bale JR, Stoll BJ, Lucas AO, editors. Improving birth outcomes: meeting the challenge in the developing world. Washington (DC): National Academies Press; 2003.
3. World Health Organization. Making Pregnancy Safer report by the secretariat. Geneva: WHO; 2000 (EB 107/26).
4. World Bank, Human Development Network. Safe motherhood and the World Bank: lessons from 10 years of experience. Washington (DC): World Bank; 1999.
5. Garner P, Kale R, Dickson R, Dans T, Salinas R. Getting research findings into practice: implementing research findings in developing countries. BMJ 1998;317:531-5.
6. Geyoushi BE, Matthews Z, Stones RW. Pathways to evidence-based reproductive healthcare in developing countries. Br J Obstet Gynaecol 2003;110:500-7.
7. Pakenham-Walsh N, Smith R. Meeting the information needs of health workers in developing countries: a new programme to coordinate and advise. BMJ 1997;314:90.
8. World Health Organization. Reproductive health library [available on the Internet and CD-ROM]. No 5. Geneva: WHO; 2002. Available from: http:// www.rhlibrary.com/
9. US Agency for International Development. Demographic and Health Surveys: Cameroon [online resource]. Calverton (MD): MeasureDHS; 1998. Available from: http:// www.measuredhs.com
10. UNICEF. State of the World's children, 2004: girls, education and development. New York (NY): UNICEF; 2003.
11. Patton MQ. Qualitative evaluation and research methods. 2nd ed. Newbury Park (CA): Sage; 1990.
12. Sofaer S. Qualitative research methods. Int J Qual Health Care 2002; 14:329-36.
13. Prendiville WJ, Elbourne D, McDonald S. Active versus expectant management in the third stage of labour. Issue 1. Oxford: Update Software 2002 (Cochrane Review).
14. Duley L, Henderson-Smart D Magnesium sulphate versus diazepam for eclampsia. Issue 1. Oxford: Update Software; 2002 (Cochrane Review).
15. Duley L, Henderson-Smart D. Magnesium sulphate versus phenytoin for eclampsia. Issue 1. Oxford: Update Software; 2002 (Cochrane Review).
16. Duley L, Gulmezoglu AM. Magnesium sulphate versus lytic cocktail for eclampsia. Issue 1. Oxford: Update Software; 2002 (Cochrane Review).
17. Crowley P. Prophylactic corticosteroids for preterm birth. Issue 1. Oxford: Update Software; 2002 (Cochrane Review).
18. Brocklehurst P, Volmink J. Antiretrovirals for reducing the risk of mother-tochild transmission of HIV infection. Issue 1. Oxford: Update Software; 2002 (Cochrane Review).
19. Smaill F, Hofmeyr GJ. Antibiotic prophylaxis for caesarean section. Issue 1. Oxford: Update Software; 2002 (Cochrane Review).
20. Smaill F. Antibiotics for asymptomatic bacteriuria in pregnancy. Issue 1. Oxford: Update Software; 2002 (Cochrane Review).
21. Lumley J, Watson L, Watson M, Bower C. Periconceptional supplementation with folate and/or multivitamins for preventing neural tube defects. Issue 1. Oxford: Update Software; 2002 (Cochrane Review).
22. Cheng L, Gülmezoglu AM, Ezcurra E, Van Look PFA. Interventions for emergency contraception. Issue 1. Oxford: Update Software; 2002 (Cochrane Review).
23. Hofmeyr GJ, Hannah ME. Planned caesarean section for term breech delivery. Issue 1. Oxford: Update Software; 2002 (Cochrane Review).
24. Hodnett ED. Caregiver support for women during childbirth. Issue 1. Oxford: Update Software; 2002 (Cochrane Review).
25. Mahomed K. Iron and folate supplementation in pregnancy. Issue 1. Oxford: Update Software; 2002 (Cochrane Review).
26. Hofmeyr GJ, Kulier R. External cephalic version for breech presentation at term. Issue 1. Oxford: Update Software; 2002 (Cochrane Review).
27. Carroli G, Belizan J. Episiotomy for vaginal birth. Issue 1. Oxford: Update Software; 2002 (Cochrane Review).
28. Dillman DA. Mail and internet surveys: the tailored design method. 2nd ed. New York (NY): John Wiley; 2000.
29. Miles MB, Huberman AM. Qualitative data analysis: an expanded sourcebook. 2nd ed. Thousand Oaks (CA): Sage; 1994.
30. Lwanga SK, Lemeshow S. Sample size determination in health studies: a practical manual. Geneva: World Health Organization; 1991.
31. Qian X, Smith H, Zhou L, Liang J, Garner P. Evidence-based obstetrics in four hospitals in China: an observational study to explore clinical practice, women's preferences and providers's views. BMC Pregnancy Childbirth [serial on the Internet] 2001;1:1. Available from: http://www.biomedcentral.com/ 1471-2393/1/1
32. World Health Organization. Coverage of selected health services for HIV/AIDS prevention and care in less developed countries in 2001. Geneva: WHO; 2002.
33. World Health Organization. WHO motherbaby package: implementing safe motherhood in countries. Geneva: WHO; 1994.
34. Meadow WL, Bell A, Sunstein CR. Statistics, not memories: what was the standard of care for administering antenatal steroids to women in preterm labor between 1985 and 2000? Obstet Gynecol. 2003;102:356-62.
35. Villar J, Carroli G, Gulmezoglu AM. The gap between evidence and practice in maternal healthcare. Int J Gynaecol Obstet 2001;75 Suppl 1:S47-54.
36. Sackett DL, Rosenberg WM, Gray JA, Haynes RB, Richardson WS. Evidencebased medicine: what it is and what it isn't. BMJ 1996;13;312:71-2.
37. Evidence-Based Medicine Working Group. Evidence-based medicine: a new approach to teaching the practice of medicine. J Am Med Assoc 1992;268:2420-5.
38. Neilson JP. Evidence-based intrapartum care: evidence from the Cochrane Library. Int J Gynaecol Obstet. 1998;62 Suppl 1:S97-102.
39. Global Health Council. Making childbirth safer through promoting evidence-based care [technical report]. Washington (DC): Global Health Council; 2002.
40. Johanson R, Lucking L. Evidence-based medicine in obstetrics. Int J Gynaecol Obstet 2000;72:179-85.
41. Sackett DL. Clinical epidemiology: what, who and whither. J Clin Epidemiol 2002;55:1161-6.
42. Page L. The backlash against evidence-based care. Birth 1996;23:191-2.
43. Grol R. Beliefs and evidence in changing clinical practice. BMJ 1997;315:418-21.
44. Grol R, Grimshaw J. From best evidence to best practice: effective implementation of change in patients' care. Lancet 2003;362:1225-30.
45. Gulmezoglu AM. Promoting standards for quality of maternal health care. Br Med Bull 2003;67:73-83.
46. Cockburn J. Adoption of evidence into practice: can change be sustainable? Med J Aust 2004;180 Suppl 6:S66-7.
(Submitted: 21 April 2005 Final revised version received: 29 August 2005 Accepted: 1 September 2005)
1 Correspondence should be sent to this author (alan.tita@obgyn.uab.edu).