INVESTIGACIÓN ORIGINAL ORIGINAL RESEARCH
Knowledge of prenatal health care among Costa Rican and Panamanian women
Conocimientos de las mujeres de Costa Rica y Panamá sobre el cuidado de la salud durante el embarazo
William Harold Guilford; Kara Elizabeth Downs; Trevor Joseph Royce
University of Virginia, Biomedical Engineering, Charlottesville, Virginia, U.S.A. Send correspondence to: W.H. Guilford, Associate Professor, University of Virginia, Biomedical Engineering, Box 800759, Charlottesville, VA 22908, U.S.A.; Tel: 434-243-2740; e-mail: email@example.com
OBJECTIVES: There is evidence that health care during pregnancy is a crucial component in ensuring a safe delivery. Because the infant mortality rate in Costa Rica is almost half the rate of Panama, the researchers tested the hypothesis that women in Costa Rica are more knowledgeable about prenatal health care than women in neighboring Panama.
METHODS: A multiple-choice survey was used to evaluate womens knowledge of prenatal care using WHO recommendations as the nominal standard. Oral surveys were administered to 320 women in Costa Rican and Panamanian health care clinics. The surveys consisted of multiple-choice questions designed to assess four specific domains of knowledge in prenatal care: nutrition, danger signs, threats from illness, and acceptable activities during pregnancy. Survey answers were scored, and significant factors in assessing womens knowledge of prenatal care were determined using analysis of variance and general linear models.
RESULTS: Costa Rican women scored higher than Panamanian women in most domains of knowledge in prenatal health care. Only country of origin and educational level were significant factors in determining knowledge of prenatal care. However, country of origin was a stronger predictor of knowledge of prenatal care than was having completed high school.
CONCLUSIONS: These data suggest that Costa Rican women are more knowledgeable about necessary prenatal care than Panamanian women, and that this difference is probably related to direct education about and promotion of prenatal care in Costa Rica. This suggests an influence of cultural health care awareness that extends beyond the previously established negative correlation between maternal educational level and infant mortality.
Keywords: Maternal health services; prenatal care; health knowledge, attitudes, practice; Costa Rica; Panama.
OBJETIVOS: El cuidado de la salud durante el embarazo es un componente crucial para garantizar un parto sin riesgo. Como la mortalidad infantil en Costa Rica es casi la mitad de la registrada en Panamá, se probó la hipótesis de que las mujeres costarricenses conocen más acerca de los cuidados durante el embarazo que las panameñas.
MÉTODOS: El conocimiento de las mujeres acerca de los cuidados de la salud durante el embarazo se evaluó mediante una encuesta, con las recomendaciones de la Organización Mundial de la Salud como estándar nominal. La encuesta se aplicó verbalmente a 320 mujeres en clínicas de Costa Rica y Panamá. Las preguntas de selección múltiple evaluaron el conocimiento en cuatro dominios específicos: nutrición, señales de alarma, amenazas por enfermedades y actividades aceptables durante el embarazo. Se asignaron puntuaciones a las respuestas. Se emplearon el análisis de varianza y modelos lineares para establecer los factores significativos que determinaron el conocimiento sobre los cuidados prenatales.
RESULTADOS: Las mujeres costarricenses tuvieron una mayor puntuación que las panameñas en la mayoría de los dominios del conocimiento sobre los cuidados de la salud durante el embarazo. Los únicos factores significativos que determinaron esos conocimientos fueron el país de origen y el nivel educacional. El país de origen fue un factor de predicción de estos conocimientos más potente que haber completado la enseñanza media.
CONCLUSIONES: Los resultados indican que las mujeres costarricenses tienen más conocimientos sobre los cuidados necesarios durante el embarazo que las panameñas y que esa diferencia puede estar relacionada con la educación directa sobre los cuidados prenatales y su promoción en Costa Rica. Esto podría indicar que hay una influencia de la cultura de concientización del cuidado de la salud que va más allá de la correlación negativa ya conocida entre la mortalidad infantil y el nivel de educación de la madre.
Palavras clave: Servicios de salud materna; atención prenatal; conocimientos, actitudes y práctica en salud; Costa Rica, Panamá.
While both Costa Rica and Panama are developing nations with limited access to health care services, the disparities in infant and maternal mortality rates are striking. Infant mortality rates have declined across Central America over the past 50 years, and a further decrease in infant mortality remains an objective in Panama and Costa Rica (1, 2). In Costa Rica the infant mortality rate is 11 per 1 000 live births compared with Panamas, where the infant mortality rate is nearly double (19 per 1 000 live births). Likewise, Panamas maternal mortality rate is over six times that of Costa Rica (160 per 100 000 live births compared with 25 per 100 000 live births) (3). The reasons for these differences are unclear, but may be related to application of prenatal care.
There is substantial evidence that health care during pregnancy is a crucial component of ensuring a safe delivery and a healthy mother and baby. The World Health Organization has published fundamental practices for a successful pregnancy, which include visiting a skilled health care worker at least four times during pregnancy, maintaining a healthy diet, knowing the signs of labor so as to seek delivery care at the appropriate time, and understanding danger signs during pregnancy (4). Women who receive prenatal care have lower rates of maternal and infant mortality, as well as better pregnancy outcomes (5), and utilization of prenatal care is correlated with higher mean birthweight and gestational age (6). Low birthweight and premature delivery have been shown to contribute to infant distress in Panama (7). Furthermore, children of mothers who did not receive prenatal care are twice as likely to die during infancy as children of mothers who received prenatal care (8).
Higher levels of general education among women are associated with improved birth outcomes (914). Indeed, the educational level of the mother is an especially significant factor in predicting infant mortality in Panama (15, 16). Whether education in general, or knowledge of prenatal care in particular, is the key factor in improving birth outcomes remains an open question. As a first step toward answering this question, we used an oral survey to evaluate womens knowledge of prenatal care in Costa Rica and Panama using the World Health Organizations recommendations as the nominal standard. We find that the country of origin is a better predictor of knowledge of prenatal care than is general educational level, though both are factors. The results of our study support our hypothesis that Costa Rican women are more knowledgeable about prenatal health care than women in Panama. This is consistent with knowledge of prenatal care contributing to the differences in infant and maternal mortality between these neighboring countries.
MATERIAL AND METHODS
In June and July of 2006, 320 women were surveyed in Costa Rica and Panama to assess their knowledge of prenatal health care. In Costa Rica, the interviews were conducted daily in six local Caja Costaricense de Seguro Social (CCSS) public health clinics in the San José and Alajuela regions. The CCSS is the Costa Rican governments social security organization which provides health care at a relatively low cost to all insured citizens. In Panama, interviews were conducted in five different rural banana plantation communities in the Bocas del Toro province. The plantation communities, known as "fincas," were within an approximate three-hour drive of the city of Changuinola. One urban clinic was held in a local church in Changuinola. The clinics where interviews were conducted were run by Global Medical Training (GMT), a medical service organization. GMT provides free health care to underserved populations in Central America and works closely with local health care providers of the Panamanian Ministry of Health.
All survey participants had come to free clinics seeking attention of a medical doctor and agreed to answer our survey while waiting. Local Spanish-speaking translators were hired to conduct the personal interviews to overcome language barriers and make the participants more comfortable. In Panama, a second translator was used to interpret for speakers of the local Ngöbe tribe dialect; 6 of the 127 Panamanians interviewed spoke this dialect. The Ngöbe are an indigenous people originating in the central mountains of Panama.
Permission to interview was granted by the directors of the facilities before the study was commenced each day. To select participants, the researchers and a translator approached individuals who appeared to be above the age of 18, without pre-selection based on pregnancy history, formal education, or overall health. If the woman was interested in participating, informed consent was recorded on audio tape and the interview, roughly 5 minutes in duration, was performed. The investigators assisted the interviewer (the hired translator) and recorded participants answers in an organized grid in handwritten field notes. Additional observations, opinions, local stories, and histories from individual conversations with the women were recorded in the field notes and on tape. No identifying information was recorded.
The research protocol was approved by the Social and Behavioral Sciences Institutional Review Board at the University of Virginia, Charlottesville, Virginia, U.S.A.
A 15-question oral survey (see the Appendix) was developed to assess participants knowledge of prenatal health care practices. The researchers also collected basic demographic data on the subject populations (Table 1). The survey consisted primarily of knowledge-based, multiple-choice questions that required no previous participant preparation. Respondents were asked to qualify the level of importance of particular prenatal care practices (e.g., a = very important, b = little importance, c = not important). "Distracter" answer choices (e.g., "headaches" as a potential danger sign) were included in each section to ensure that participants were focused and answering after critical thought. Simple, fifth-grade level vocabulary was used and, because of the low literacy rates characteristic of both Costa Rica and Panama, an oral survey design was chosen. Standardized question format and answer choices were translated into Spanish by a local Costa Rican, so that consistent wording was used by the translators.
The knowledge to be assessed in the survey was derived from several sources. The World Health Organizations critical criteria for care during pregnancy were used as the nominal standard. In addition, the investigators consulted an obstetrics and gynecology specialist (personal communication, Y. Newberry, Family Nurse Practitioner, University of Virginia, Charlottesville, Virginia; April 2006), the American College of Obstetrics and Gynecology (ACOG) guidelines on prenatal nutrition (17), and the University of Michigan guidelines for prenatal clinical care (18).
The prenatal care information evaluated in the survey was organized into five major categories: diet, danger signs, illnesses, positive and negative activities, and labor signs. These aspects of prenatal care are most effectively encouraged and monitored through meetings with a skilled health care worker before, during, and after pregnancy. WHO advises that women should visit a health center at least four times during pregnancy (4); however, ACOG recommends at least twelve routine visits (17). A common theme encountered in global health is the dilemma that arises when the standards of developed countries exceed those possible in the developing world. Therefore, we designed the survey to take into account the potentially limited access to health care of the subject population. This was accomplished by assessing the minimum level of knowledge and recommendations for a safe, healthy pregnancy.
The dietary section of the survey was designed based on the recommendation for pregnant women to maintain a healthy, well-balanced diet (18). Foods particularly high in nutrients, such as spinach, are stressed because the Recommended Daily Allowance (RDA) of most nutrients increases during pregnancy (17). The survey focused on foods that would meet these requirements and were available to the local population.
WHO also recommends that pregnant women learn and recognize danger signs, so that they will know when to seek the aid of a skilled health care worker. Therefore, potentially dangerous conditions, as well as illnesses that could have adverse effects on both mother and unborn baby, were included in the survey. Women who are informed of dangerous signs and illnesses that would require special attention will ideally be able to prevent negative outcomes for their babies and themselves.
Pregnancy differs from other "conditions" in which consistent medical care is advised because it is not an illness. It is recommended that women maintain their respective normal lifestyles during pregnancy, but avoid harmful activities such as using tobacco, alcohol, and medications not approved by a health care worker (over-the-counter or illicit), and participating in overly strenuous activities. Practices that can positively affect the health of mother and baby include taking multivitamins, folic acid and iron supplements, and drinking plenty of purified water (5). A number of these positive and negative activities were evaluated in the survey.
Finally, recognizing signs of labor is a vital part of preventing adverse outcomes in childbirth. The WHO recommends that women be able to identify such signs, including painful contractions, broken water, or bloody and sticky discharge. The corresponding section of the survey included five potentially significant indications of labor.
Thirteen of the 15 questions on the survey were standard, multiple-choice format. The remaining two questions were about the participants age, and an open-ended commentary on the study.
As interviews were completed, each participants responses were recorded in an Excel database. The survey answers were converted into a numerical scoring system ranging from 0 to 2, indicating their degree of correspondence with best practices. 0 indicated an incorrect answer, 2 indicated a correct answer, and 1 indicated answer choices that could apply in some situations (e.g., a "maybe" response). After assigning each response a score, the average score of each category for each individual was calculated. This technique assigned each category the same statistical weight while the number of questions in each category varied. Statistics were calculated in Microsoft Excel (Microsoft Inc., Redmond, Washington, U.S.A., 2003) and SPSS software (SPSS Inc., Chicago, Illinois, 2006). Specific statistical tests are noted in the results. P-values less than 0.05 were considered significant.
A total of 320 women were interviewed; 193 were Costa Rican (60%) and 127 were Panamanian (40%) (Table 1). The mean age of Costa Rican respondents was 33, ranging from 18 to 74 years old. The mean age of Panamanian subjects was 31 with a range from 18 to 61 years old. Women under the age of 18 were excluded from the participant pool. Neither the age nor the education level of participants differed significantly between the two countries (analysis of variance [ANOVA] and z-test); 34.2% of the Costa Rican and 40.1% of the Panamanian subjects had finished high school.
The economic difference between our two sample populations was much larger than might be expected, considering the similarity in their education levels. Of the Costa Ricans, 29% fell into our "high income" bracket, whereas only 6.3% of Panamanians were categorized as having "high income." Whether or not the respondents pay for their water was another economic measure used. This information is also an indicator of the quality of water supplied to the respondents households. Of the Costa Ricans surveyed, 93.2% pay for their water, while only 50.4% of the Panamanians do.
The numeric results of tests of knowledge in both countries were normally distributed, as determined by Lilliefors test. Mean scores for each category are given in Table 2. The overall knowledge of prenatal care (un-weighted mean of categories) differed significantly between Panama and Costa Rica, suggesting a large disparity in maternal knowledge between the two countries. In every category except "signs of labor," the mean scores of Costa Ricans are higher than Panamanians as determined by z-tests and the Mann-Whitney test (Table 3). This suggests that the Costa Ricans and Panamanians interviewed are similarly knowledgeable of signs of labor. In a separate question (question 13; see Appendix) there was no significant difference between Costa Rican and Panamanian women in their knowledge of the best place to give birth (i.e., hospital, local clinic, or at home).
Univariate ANOVA in a general linear model (GLM) was used to determine what factors are most predictive of overall knowledge of prenatal care. The results are given in Table 4. Only country and educational level were significant factors in determining overall knowledge of prenatal care. This is in general agreement with other researchers who found that educational level is a significant predictor of infant mortality in the two countries (15), but especially significant in Panama (15,16). Neither income, paying for water (an indicator of both income and clean water availability), nor age were significant factors. When the countries are considered separately, education emerges as the only significant factor among Costa Rican women, whereas there were no significant factors differentiating Panamanian women (data not shown) in terms of their knowledge of prenatal care. Considered in those terms, the Panamanian women interviewed more closely resemble Costa Rican women in the sample who had not completed high school.
To further understand educational level and country of origin as factors in determining knowledge of prenatal care, a multivariate GLM was utilized. The significance of these factors in the GLM is given in Table 5. While both country and educational level are significant factors, the country of origin has a larger effect than having graduated from high school. The low partial eta-squared (ηp2) suggests that education is a weak factor in the GLM compared to country. We can further dissect the relationship between these factors and individual measures of knowledge (see Table 6). Education is a significant factor in every area of knowledge except good and bad practices during pregnancy. Country is a significant factor in every area of knowledge except signs of labor.
The data are consistent with our hypothesis that women in Costa Rica are more knowledgeable about prenatal health care than women in Panama. Since Costa Rica has lower rates of infant and maternal mortality, and since it is accepted that prenatal care has positive impacts on pregnancy outcomes, there is reason to believe that improving knowledge of prenatal care will improve both maternal and infant health. While we cannot make a direct correlation between higher rates of mortality and lower levels of knowledge, it is important to recognize the potentially influential aspects of prenatal health care awareness for the health and well-being of both mothers and children. Our data suggest that a solid starting point for reducing the number of infant and maternal deaths may be the education of women about basic prenatal care practices. Bhutta (9) pointed out that while such approaches have been tried in industrialized countries, little is known about its efficacy in either industrialized or developing nations. Our study, while offering no experimental interventions, does predict an impact of prenatal health care education.
We encountered considerably greater interest and awareness about prenatal care in Costa Rica than in Panama. While almost every participant in both countries was willing to participate, in some cases the Costa Rican women approached us showing interest in participating and often requested feedback on their scores. Several participants commented that they would remember what they had learned for future use. Willingness to learn and a general positive attitude toward health care seem to be closely related to womens interest in prenatal care.
While culture certainly influences these opinions, access to health care providers plays an invaluable role. Both Costa Rica and Panama offer universal social insurance, but access is more limited in Panama (19). In Costa Rica, the CCSS public health facilities appear to be effectively reaching a wide range of their citizens. This is due in part to the decentralized nature of the health care system, in which hospitals and large clinics are located in central areas, and smaller clinics, often one or two rooms in size (locally known as EBAIS or basic care health teams), are located in the rural areas. In this way, secluded villages and women without access to transportation are able to receive health care. In Panama, the Ministry of Health has organized primary, secondary, and tertiary levels of care, but with less extensive management (20).
The small number of Ngöbe participants (6 of 127 Panamanians interviewed) within our data set does not affect the results. The small Ngöbe sample does not differ significantly in any of the knowledge categories from the rest of the Panamanians surveyed, though in absolute terms they scored somewhat lower in every knowledge category. It has been suggested by the United Nations Population Fund (UNFPA) that in the rural, mountainous regions of Panama, where the indigenous Ngöbe reside, lack of transportation to health care facilities and rugged lifestyle contribute to complications with childbirth (21). Regardless, the authors believe that lack of access to medical services does not necessarily predict a lower level of prenatal care knowledge, though additional studies will be needed to test this hypothesis.
This study complements previous work suggesting that higher levels of general education among women are associated with improved birth outcomes (9, 1114). We find that maternal educational level is a significant factor in determining the knowledge of prenatal care even in the face of large disparities in economic status and urban versus rural populations. However, the country of residence outweighs educational level as a factor in predicting knowledge. This is reflected in the statistically equal fraction of Costa Rican and Panamanian participants who had completed high school, yet they displayed a significant gap in knowledge about prenatal care. This suggests an indirect effect of education on knowledge of prenatal care. For example, an educated local populace can give rise to culturally or socially imbued knowledge of prenatal care (22). Educated women may also be more comfortable interacting with a modern health care system (13). It is equally possible that Costa Ricas government is working successfully to inform women about prenatal care.
An interesting observation is that despite a significant knowledge gap between the two countries in many areas, all women, regardless of country of residence, appear to be equally knowledgeable about signs of labor. This may be due to knowledge of labor being inherent, being based on personal experience, or being passed between family or community members. Furthermore, according to surveys done in 1998, 90% or more of all births in Panama and Costa Rica were assisted by a skilled birth attendant, suggesting great labor awareness in both countries (23). It should also be noted that knowledge of the signs of labor, while beneficial to care, is not necessarily indicative of being better informed about prenatal care.
Recently the CCSS of Costa Rica made efforts to promote health care and prevention strategies as part of its services (20). For example, in a clinic in Higuito, Costa Rica, the nurses publicized a series of free weekly information sessions for pregnant women. We observed posters displaying information with recommended practices that were organized into similar categories as those in our survey. Pamphlets with information were often readily available at the clinic and a billboard on the side of a busy street encouraged mothers to care for themselves during pregnancy. This approach has been shown to be effective for promoting postnatal care in Nepal (24).
Limitations of the study. It is evident that citizens with higher incomes (and perhaps a greater degree of maternal knowledge) see doctors in private clinics that are not paid by the government (personal communication, G.C. Galliano, M.D., Global Medical Training, 20 June 2006). We therefore chose to obtain our sample population from public clinics, because this is where all citizens, regardless of socioeconomic status, can receive equal care. This possibly biased our sample toward populations in a lower economic bracket. In addition, it should be considered that all women were interviewed in a medical setting. This would suggest that they have a predisposition to health care and a desire to keep both themselves and their children healthy, and may, therefore, possess a greater degree of maternal knowledge.
We reported the education level of participants based on whether they had graduated from high school. While our survey instrument was not designed to assess educational level in more detail, standard levels of education seemed to be at least through primary school; attending high school was less common and considered something of a privilege. Very few participants mentioned that they had attended college. Maternal education for women who have completed primary school but not high school is correlated with reduced infant mortality (916). However, it remains to be determined whether lower educational levels (e.g., women who have completed primary school versus those who have not) affect knowledge of perinatal care.
This study does not demonstrate a causal link between knowledge of prenatal care and improved pregnancy outcomes. To more directly assess such a relationship it would be necessary to question each subject about her pregnancy outcome. We chose not to question women on their childbirth experiences for two reasons. First, an estimated six instances of infant mortality would be expected in our sample size, making correlative statistics impossible. A much larger sample size would be required. The second reason was concern for the participants social and mental welfare when responding to such questions could be traumatic. Now that baseline relationships have been established, such a study may be justifiable.
We conclude that Costa Rican women are more knowledgeable about necessary prenatal care than Panamanian women, and that this difference is probably related to direct education about and promotion of prenatal care in Costa Rica. While additional studies will be needed to separate prenatal health care education from other activities that affect pregnancy outcomes, our data suggest that focused educational efforts may prove a straightforward and relatively inexpensive approach to decreasing perinatal mortality. Peer education subsequent to such government interventions may benefit local communities as well as entire nations with minimal resources.
Acknowledgments. The researchers would like to thank Yvonne Newberry, University of Virginia, for her dedication and advice in the development of the project. We would also like to thank Karen Schmidt (University of Virginia) and the Global Medical Training Staff, including Santiago Mora, Gian-Carlo Galliano, Wil Johnson, Curtis Larsen, and Sondra Elizondro. We are most appreciative of the generous support of the Stull family and the Harrison Special Collection Institute of the University of Virginia.
1. Schneider MC, Castillo-Salgado C, Loyola-Elizondo E, Bacallao J, Mujica OJ, Vidaurre M, et al. Trends in infant mortality inequalities in the Americas: 19551995. J Epidemiol Community Health. 2002;56(7):53841.
2. Guzman JM. Trends in socio-economic differentials in infant mortality in selected Latin American countries. In: Ruzicka L, Wunsch G, Kane P, eds. Differential mortality: methodological issues and biosocial factors. Oxford: Oxford University Press; 1995. Pp. 13144.
3. World Health Organization (WHO). Statistical Information Systems. Core Health Indicators [Internet site]. Available from: http://www.who.int/whosis/en. Accessed 5 December 2005.
4. World Health Organization (WHO). Pregnancy, childbirth, postpartum and newborn care: a guide for essential practice. Geneva: WHO; 2006.
5. Lockwood CJ. The initial prenatal assessment and routine prenatal care. UpToDate [Internet site]. Available from: http://patients.upto date.com/topic.asp?file=antenatl/2937. Accessed 18 March 2007.
6. Alexander GR, Cornely DA. Prenatal care utilization: its measurement and relationship to pregnancy outcome. Am J Prev Med. 1987; 3(5):24353.
7. Moreno MT, Vargas S, Poveda R, Sáez-Llorens X. Neonatal sepsis and meningitis in a developing Latin American country. Pediatr Infect Dis J. 1994; 13(6):51620.
8. Hong R, Ruiz-Beltran M. Impact of prenatal care on infant survival in Bangladesh. Matern Child Health J. 2007;11(2):199206.
9. Bhutta ZA, Darmstadt GL, Hasan BS, Haws RA. Community-based interventions for improving perinatal and neonatal health outcomes in developing countries: a review of the evidence. Pediatrics. 2005;115(2 Suppl): 519617.
10. Hojman DE. Economic and other determinants of infant and child mortality in small developing countries: the case of Central America and the Caribbean. Appl Econ. 1996; 28(3):28190.
11. Buor D. Mothers education and childhood mortality in Ghana. Health Policy. 2003;64(3): 297309.
12. Hobcraft J. Womens education, child welfare and child survival: a review of the evidence. Health Transit Rev. 1993; 3(2):15975.
13. Joshi AR. Maternal schooling and child health: preliminary analysis of the intervening mechanisms in rural Nepal. Health Transit Rev. 1994;4(1):128.
14. Ricci JA, Becker S. Risk factors for wasting and stunting among children in Metro Cebu, Philippines. Am J Clin Nutr. 1996;63(6):96675.
15. Hobcraft JN, McDonald JW, Rutstein SO. Socio-economic factors in infant and child-mortality: a cross-national comparison. Popul Stud (Camb).1984;38(2):193223.
16. Frey RS, Field C. The determinants of infant mortality in the less developed countries: a cross-national test of five theories. Soc Indic Res. 2000; 52(3):21534.
17. American College of Obstetrics and Gynecology (ACOG). Nutrition during pregnancy. Washington, D.C.: ACOG; 2007. (Patient education series #AP001).
18. Zoschnick LB, Brackbill EL, Schumacher R, Green LA, Harrison RV. Prenatal care guidelines. Ann Arbor: University of Michigan Health System; 2006.
19. Mesa-Lago C. Social security in Latin America: pension and health care reforms in the last quarter century. Lat Am Res Rev. 2007;42(2): 181201.
20. Pan American Health Organization. Basic country health profiles for the Americas [Internet site]. Available from: http://www. paho.org/English/DD/AIS/cp_index.htm. Accessed 28 May 2007.
21. Olfarnes, T. Risking death to give life in Panamas tropical forest [Internet site]. UNFPA News, 2 April 2007. Available from: http:// www.unfpa.org/news/news.cfm?ID=949#. Accessed 15 November 2007.
22. Stephenson R, Matthews Z. Maternal heath-care service use among rural-urban migrants in Mumbai, India. Asia Pac Popul J. 2004; 19(1):3960.
23. AbouZahr C, Wardlaw T. Maternal mortality at the end of a decade: signs of progress? Bull World Health Organ. 2001;79(6):5618.
24. Allen CW, Jeffery H. Implementation and evaluation of a neonatal educational program in rural Nepal. J Trop Pediatr. 2006;52(3): 21822.
Manuscript received on 13 July 2007.
Revised version accepted for publication on 11 January 2008.