Sociodemographic characteristics associated with the utilization of maternal health services in Cambodia

Background Cambodia is a Southeast Asian country and has one the highest rates of maternal and child mortality with inadequate use of maternal healthcare services in the region. The present study aimed to analyse the progress made in terms of using maternal healthcare services since 2000. Methods Two rounds of Demographic and Health Surveys (DHS 2000 and DHS 2014) were used in the study. Sample population consisted 11,961 women aged between 15 and 49 years. The outcome measures were: Timing of first antenatal care (ANC) attendance, adequacy of ANC attendance, place of delivery and postnatal checkup. WHO guidelines were used to set the cut-off/define these measures. Data were analyzed in Stata version 14 using descriptive and multivariate regression analyses. Results Findings indicated that the overall prevalence of making the first ANC visit in the first trimester was 64.19% [95%CI = 62.22,66.11], and that of having at least four ANC visits was 43.80% [95%CI = 41.89,45.73]. Prevalence of health facility delivery was 48.76% [46.62,50.90] and that of postnatal checkup was 71.14% [95%CI = 69.21,73.01]. Between 2000 and 2014, the percentage of timely and adequate use of ANC increased by respectively 61.8 and 65.3%, while that of health facility delivery and postnatal care increased by respectively 74.5 and 43.9%. Important demographic, socioeconomic and geographic disparities were observed in the utilization of ANC, health facility delivery and postnatal care services. Urban residency, having better educational status, white collar job, access to electronic media showed positive association, whereas higher parity (having > 2 children) and unwanted pregnancy showed negative association with the use of maternal healthcare services. Having at least four ANC visits was associated with significantly increased higher odds of using health facility delivery and postnatal care. Conclusion There has a been a remarkable increase in the prevalence of women who are using the maternal healthcare services since 2000. The current findings provide important insights regarding the sociodemographic factors associated with the utilization of maternal health services in Cambodia that could contribute to evidence-based health policy making and designing intervention programs.


Background
Cambodia is low-lying country in the Southeast Asian region, and has a population of over 16 million (as of 2019). Since obtaining independence in 1953, the country has made significant improvements in terms of reducing the percentage of absolute poverty, malnutrition, and improving the key public health indicators such as maternal and child mortality rates. Between 2000 and 2014, Cambodia managed to cut the rate of neonatal mortality by 46% [1] in parallel with an equally impressive gain in reducing maternal mortality (488 in 2000 to 160 in 2017) [2]. Despite these major advancements, the country still ranks among the least developed countries and faces critical challenges for meeting the millennium and sustainable development goals.
In recent years, healthcare system in Cambodia has implemented nationwide initiatives to promote reproductive, maternal, newborn and child health. A continuum of essential medical services such as antenatal care, skilled childbirth, and postnatal care services has been put in place in order to provide lifesaving medical services for pregnant women and their newborn. ANC services include a wide range of interventions designed to minimize obstetric complications and ensure a healthy pregnancy outcome. These services generally include anthropometric tests, blood and urine analysis, nutritional and psychological counselling, vaccination, dietary supplementation, informing the mother regarding the danger signs, mental and financial preparedness for childbirth, and importance of using skilled birthing services [3][4][5][6][7]. A study involving 69 low-middle countries found that even a single ANC visit could reduce the probability of neonatal and infant mortality respectively 1.04% points and 1.07% points [3]. In addition to routine ANC, WHO also emphasizes on choosing professional childbirth services and postnatal checkup (PNC) which are crucial for providing emergency obstetric services and avoiding the risks of prolonged labour, stillbirth, obstetric and postpartum hemorrhage [8][9][10]. These services are highly cost-effective and their proper utilization is regarded as a central strategy to reducing maternal and child mortality in all low-middle-income countries [11]. WHO recommends attending at least four antenatal visits (which was later revised to eight) during the course of the pregnancy with the first visit taking place no later than the 12 weeks from conception. Nonetheless, making these services universally available and ensuring their optimum use in an equitable manner is an extremely challenging task, especially for resource-poor countries where medical infrastructure and skilled professionals are often shockingly scarce. Geographic remoteness, high out-of-pocket costs of care, absence of skilled care provider and medium for health communication pose significant challenges for improving the use of ANC services particularly among the disadvantaged communities [12,13]. Maternal healthcare services (MHS) such as ANC are provided through primary care clinics and community health centers in the urban and rural areas. Cambodia has been experiencing rising socioeconomic standards among women driven by higher school completion rate and engagement in labour force. Agriculture is the major economic activity in the rural areas, while in the urban areas women are participating in service and manufacturing industries. Improved socioeconomic situation constitute a key determinant of healthcare utilisation among women. Nonetheless, a large segment of the population still faces significant barriers in accessing basic health services. Addressing the barriers is crucial for promoting public health and meeting the national and international development goals targeted to reducing maternal and child mortality e.g. Millennium and Sustainable Development Goals.
ANC service in Cambodia are mainly provided through the primacy care facilities, however, midwives and community health workers also play vital roles especially in the remote areas with limited access to service [13,14]. Large demographic and regional gaps remain in the provision and utilisation of these services as the healthcare system suffers from infrastructural and manpower crises. Understanding the population sub-groups who are being deprived from these services is an important step to scaling up the existing maternal health programs and making new evidence-based policies. In this regard, the present study attempts to provide an assessment of the progress made in terms of maternal healthcare services since 2000. We used two rounds of Demographic and Health Surveys (DHS 2000 and DHS 2014) that collected data on various demographic, socioeconomic and healthcare related indicators among women of childbearing age (15-49 years). DHS data for Cambodia were studied previously to identify the inequalities in the use of ANC and childbirth services. However, currently there is no study covering all the components of pre-(timing and frequency of ANC) and postpartum services (childbirth and postnatal care).

Data source
Data for this study were secondary and obtained from the Cambodia Demographic and Health Survey 2014 and 2000. The implementing organization was National Institute of Statistics (NIS -Ministry of Planning) in collaboration with the Ministry of Health. Fieldwork for the latest survey lasted from June 2014 to December 2014, and the previous one from July 2010 to January 2011. The survey received technical assistance by ICF provided through The DHS Program, a project funded by the United States Agency for International Development. This is a nationally-representative project that surveys adult men, women (15-49 years) and children under 5 years of age. The survey is cross-sectional in nature and collects data using a standard questionnaire which comprises questions covering demographic, socioeconomic, knowledge of reproductive health, and healthcare use related topics. Sampling techniques included multistage cluster approach that involved systematic selection of the geographic divisions of the country, and then the selection of households from those divisions. Detailed description of these sampling methods was published elsewhere: National Institute of Statistics, Directorate General for Health, and ICF International, 2015 [15].

Data analysis
Data analysis consisted of descriptive statistics, chisquared bivariate tests and multivariate analyses using Stata version 14. Firstly, datasets for 2000 and 2014 were merged and then descriptive statistics were used to calculate the prevalence of maternal healthcare utilisation for both surveys. This was followed by binary regression analysis (logit link) to measure the adjusted association between the outcome and independent variables. For multivariate analysis, only the variables that showed significance at p < 0.1 in the bivariate analysis were selected for multivariate analysis. For each of the four outcome variables, three sets of regression models were run: overall sample, urban and rural. The urban-rural stratification was done given the growing recognition of the regional difference in healthcare seeking and importance in designing intervention measures. Results of regression models were presented as odds ratios and 95% confidence intervals. Statistical significance was assumed at p < 0.05. We used Variance Inflation Factor in Stata to diagnose multicollinearity following the multivariate analyses. We did not find any such multicollinearity to report as the Variance Inflation Factor values for all the models were far below the cut-off of 10. Table 1 summarises the prevalence of utilization of the four types of services. Sample population consisted of 11, 961 women aged 15-49 years. Descriptive results indicated that a greater proportion of the women who used these services were generally aged between 25 to 34 years, rural residents, had primary level education, from higher wealth index households, male-headed households, employed in blue collar jobs, made healthcare decisions by themselves, had access to electronic media, reported last child as wanted, and had 1-2 children. Figure 1 shows that between 2000 and 2014, the percentage of timely and adequate use of ANC increased by respectively 61.8 and 65.3%, while that of health facility delivery and postnatal care increased by respectively 74.5 and 43.9%.

Results
Sociodemographic inequalities in timely and adequate use of ANC services were presented in Table 2 20.67] increased substantially in 2014. The odds of these outcomes increased among women in the higher age groups, except for those aged > 44 years. Rural residence showed a significantly negative association with adequate use of ANC services [Odds ratio = 0.67, 95%CI = 0.55,0.82]. Increased odds ratios were also observed among women who had secondary/higher education, and the same was true when the husband had secondary/higher education as well. Wealth status didn't show any significant association with early initiation of ANC, however, women in the middle [Odds ratio = 1.30, 95%CI = 1.05,1.62] and richer [Odds ratio = 1.49, 95%CI = 1.18,1.88] wealth quintiles were more likely to have adequate ANC visits. Women who were employed in white collar profession had significantly higher odds of timely and adequate use of ANC. Having access to electronic media was associated higher odds of making adequate ANC visits [Odds ratio = 1.45, 95%CI = 1.22,1.73]. Having unwanted child and higher parity (> 2 children) showed a negative association with both timely and adequate use of ANC services. More details were shown in Table 2.
Sociodemographic inequalities in the use of professional childbirth and PNC services were presented in Table 3. Similar to using ANC services, the odds of professional childbirth [Odds ratio = 17.89, 95%CI = 14.55,  Table 3.

Discussions
The objective of this study was to assess the situation of maternal healthcare service utilization in Cambodia. Previous studies have highlighted that services such as timing and frequency of using antenatal care, place of delivery, and postnatal checkup are vital components of MHS. Promoting the use of MHS lies at the center of attaining maternal and child health related Sustainable Development Goals. From the descriptive analysis we found that little less than two-third of the women made their first ANC visit in the first trimester, while more than two-fifth attended four or more ANC visits. Despite a relatively lower ANC attendance, we observed a relatively higher percentage of health facility delivery. The prevalence of postnatal checkup was even higher which stood at 71.14%. Although a large percentage of women are still not being able to access these life-saving healthcare services, there has been a considerable improvement in the percentage of women who used these services since 2000. Comparing the latest statistics with those from 2000 suggests that the percentage of timely and adequate use of ANC increased by respectively 61.8 and 65.3%, while that of health facility delivery and    The positive trend marks a major healthcare success for a low-income country like Cambodia, and provides a good example for other countries that are still lagging behind the international targets. Continued efforts will be required to further promote this success with the goal of reaching universal coverage of maternal healthcare services in Cambodia. Further analyses were performed to identify the sociodemographic factors that are associated with the use of these services. Higher age, urban residence, better educational status, household wealth quintile (with some exceptions), employment status, access to electronic media, relatively lower parity and not having unwanted pregnancy were found to have a protective effect on the use of the healthcare services. The patterns behind these associations are in line with previous studies, and reflect the role of socioeconomic status which acts as an enabling factor for accessing healthcare, and overall health and well-being of individuals [18-20, 25, 26]. The positive role of educational status on the use of the services of the husband was particularly interesting. The absence of correlation between household wealth status with timely initiation of ANC and PNC checkup was a potentially contradictory one, although wealth status did show a significant association with having adequate ANC visits and using health facility delivery. Household financial well-being is a strong predictor of maintaining a healthy pregnancy and accessing essential medical care to ensure a successful termination of pregnancy, and health of the mother and her newborn [27][28][29][30]. This finding indicates that wealth inequality is an important factor in the use mater healthcare services. In low-income countries like Cambodia, out-of-pocket expenditure can pose serious constraints to availing life-saving medical services [5,31,32]. Reducing socioeconomic inequalities through patient-focused financial programs such as removal of user fees can prove beneficial to promoting the use of healthcare services and contribute to lower rates of maternal and child mortality.
Having access to electronic media showed a positive association with using adequate ANC and PNC services. Health communication through electronic media is an increasingly popular strategy to improve health knowledge and self-efficacy, which eventually contribute to more effective use of healthcare services and better population health [5,[33][34][35][36]. Regarding parity, women who had more than > 2 children were less likely to use healthcare services compared with those who had 1-2 children. The exact explanation behind this association can vary depending on individual circumstances, however, it is likely that women who experienced several pregnancies consider themselves being aware of pregnancy related health risks, and thus made healthcare visits less frequently [21,37,38]. Another explanation might be that the higher costs associated with having higher number of children may prevent women from seeking care. Women who reported their last pregnancy as unwanted were significantly less likely to access care. Unwanted pregnancy is a complex issue and has been shown to have negative health outcomes for both mother and children in previous studies [17,[39][40][41]. Healthcare strategies that can help prevent unwanted pregnancies include better provision of family planning products and counselling, especially among newly married couples. Effective use of family planning services can also increase reproductive health awareness, and thereby promote the care-seeking behaviour. Lastly, we found that women who attended four or more ANC visits were more likely to deliver at a health facility and take PNC checkup. In primary care settings, ANC services can greatly increase women's exposure to health counselling and discussion with care providers, which enable them to make better healthcare decisions. From this view, improving ANC attendance can result in higher percentage of health facility delivery and PNC checkup, two key strategies for preventing maternal and child mortalities [42][43][44]. This study has several strengths and limitations to report. This study used large nationally-representative data from Cambodia Demographic and Health Survey, which is a nationally-representative survey that collects data on various domains including demographic, socioeconomic, healthcare related variables and adult men and women. The latest data were collected in 2014, and therefore may not represent the recent situation. Although the data are little old, they provide a comparative picture of the trend in maternal healthcare utlisation during the 2000-2014 period. As mentioned earlier, this was a secondary study which limited the choice of the variables or sample characteristics. For this reason, some critical variables such as cultural and infrastructure related factors were not included in the analysis. There were also no data to assess the quality of the services which are important determinants of healthcare seeking behaviour. Lastly, the surveys were cross-sectional and hence the associations are not guaranteed to establish any causal relationship. Despite these limitations, the present study provides important insights regarding the prevalence and predictors of using reproductive services which can have practical implications for reproductive health programs in the country. Further research is necessary to explore the quality of the reproductive health services in addition to merely increasing the access, to ensure sustained success in terms of maternal and child health outcomes.

Conclusions
In this study we assessed recent progress in maternal healthcare use as well as their predictors in Cambodia. Findings indicate that little less than two-third of the women made their first ANC visit in the first trimester, more than two-fifth attended four or more ANC visits, more than half of the deliveries took place at respondent's home, and less three quarter had PNC checkup.
Year by year analysis revealed a significant improvement in the use the maternal healthcare services. Nonetheless, considerable disparities still exist in the utilisation of these services. Based on the findings, its suggestible that investing on women's socioeconomic status, access to electronic media, and interventions targeting to reducing higher parity and unintended pregnancies may contribute to better use of maternal health services in Cambodia.