DEVELOPMENT, PILOTING AND EVALUATION OF A FRAMEWORK FOR INCREASING UTILIZATION OF SKILLED MATERNAL HEALTHCARE SERVICES: STUDY IN SIAYA COUNTY Obinge

Obinge Elizabeth Omondi 1, 3 , Fred A. Amimo 1 and Asito Stephen Amolo 2 . 1. Department of Public Health, School of Health Sciences, Jaramogi Oginga Odinga University of Science and Technology. 2. Department of Biological Sciences, School of Biological and Physical Sciences, Jaramogi Oginga Odinga University of Science and Technology. 3. Department of Health, County Government of Siaya. ...................................................................................................................... Manuscript Info Abstract ......................... ........................................................................ Manuscript History

Utilization of skilled maternal healthcare services during pregnancy, delivery and post delivery can reduce maternal deaths from preventable conditions. However, despite the usefulness of utilizing the skilled maternal healthcare services, utilization has remained low in Siaya County while women continue to die from preventable conditions. The aim of this study was to identify the correlates of utilization of skilled maternal healthcare services in Siaya County and consequently develop, pilot and evaluate a context specific framework for increasing utilization of skilled maternal healthcare services in Siaya County. Method:The study followed a three (3) phased approach, whereby baseline data was first collected from the study participants to ascertain their past maternal health services utilization experience ( antenatal, delivery and postnatal care services), followed by development and piloting of the framework which was informed by the findings of the baseline, and finally evaluation of the effectiveness of the framework. The framework was piloted through a longitudinal study involving 517 pregnant mothers with previous delivery experience from selected community health units in Siaya County. Statistical package for social sciences (SPSS) version 22 was used to analyze both descriptive and inferential statistics.
Result: Indicate that both community and health facility factors affect utilization of skilled maternal healthcare services. Compared to the baseline, 4 th Antenatal visits increased from 58.7% to 63.9%, skilled deliveries from 55.7% to 64.6% and postnatal care services from 23.6% to 47.4%. Knowledge on available maternal healthcare services and danger signs also increased by 28.6% and 26.4% respectively. Of note is that compared to the baseline, continuum of care was also enhanced as the number of households visited by community health volunteers (CHVs) increased from 83.7% to 93.5%. an indication that there was increase in uptake of skilled maternal healthcare services. Conclusion: The framework was effective in increasing utilization of skilled maternal healthcare services in Siaya County and will be useful in providing guidance on designing and integrating context specific 755 Kenya is one of the Countries with high maternal mortality levels approximated at 495 per 100,000 live births (Kenya National Bureau of Statistics et al., 2015b), well above the MDG target of 147 per 100,000 by 2015 (Statistics, 2013). Kenya was rated among the top ten countries that contributed to 58% of the global maternal deaths reported in 2013 (U. WHO, UNFPA, The World Bank and the United Nations Population Division., 2014), with over 6,000 women dying every year due to pregnancy related conditions despite the launch of Safe Motherhood campaigns twenty years ago (Macro, 2010;Starrs, 2006). In addition, it is estimated that another 20-30 women suffer serious injury or disability due to complications during pregnancy or delivery in Kenya (Human Rights., 2012). These high rates have persisted despite improvements in other health indicators over the past decades (Health, 2014). This problem is partly due to lack of access to skilled maternal health services, including ante-natal care, skilled delivery, and post-natal services (Health, 2014). Although there has been increased growth of health sector infrastructure, many women still face various barriers to accessing skilled maternal health services (Bourbonnais, Kenya, 2012). In Kenya there are major gaps and challenges associated with supply-side and demand-side factors leading to geographical disparities in coverage of health services between counties, urban and rural residents and different population groups(Kenya National Bureau of Statistics et al., 2015a). In attempt to address the identified challenges affecting uptake of skilled maternal healthcare services, the government of Kenya introduced community health strategy to empower individual and families to take responsibility for their health (Health, 2006) and also introduced a policy of free maternity services in all public facilities (Bourbonnais, 2013). As recommended by Kerber et al., 2007; the continuum of care for maternal, neonatal, and child health requires access to care provided by families and communities, by outpatient and outreach services, and by clinical services throughout the lifecycle, including adolescence, pregnancy, childbirth, the postnatal period, and childhood. Saving lives depends on high coverage and quality of integrated service-delivery packages throughout the continuum, with functional linkages between levels of care in the health system and between service-delivery packages, so that the care provided at each time and place contributes to the effectiveness of all the linked packages (Kerber et al., 2007).
While it is acknowledged that most causes of maternal deaths could be prevented through utilization of skilled maternal healthcare services (Graham et al., 2012;Wang et al., 2010;WHO et al., 2004); utilization of skilled maternal healthcare services have remained low in Kenya especially in counties like Siaya County despite massive investments on interventions to increase uptake of these services while mothers continue to die from preventable causes (DHIS 2014). Siaya County in Kenya is one of the counties with high MMR of 691 deaths for every 100,000 live births that is above the national average of 495 deaths for every 100,000 live births with most of these deaths occurring at home (MICS4 2011). Previous studies and reports in Siaya County associated the low uptake of skilled maternal healthcare services to poor communication between health workers and women, socio cultural beliefs, distance to health facilities and quality of ANC services (Pell et al., 2011: Bottleneck analysis 2012; Siaya investment case for MNCH 2013). However despite attempts to address the identified barriers, uptake of skilled maternal healthcare services has remained low (DHIS, 2014). The high maternal mortality ratio in Siaya County has been mainly attributed to low utilization of skilled maternal healthcare services (DHIS, 2014). Data indicate that the proportion of women attending at least one ANC visit are 85%, those attending four ANC visits are 45%, 52% receive skilled care during delivery while only 30% receive postnatal care within two days after delivery (DHIS, 2014). Different policies, frameworks and strategies have emerged to respond to the growing concerns on low utilization of skilled maternal healthcare services coupled with high maternal mortality levels in developing countries (Aday, 1974;Andersen, 1995 The use of community health workers in creating demand for skilled maternal health care services in Siaya County began in 2006 after the launch of the second Kenya national health sector strategic plan (KNHSSP 11). .However despite all the effort, Counties like Siaya County continue to have high maternal deaths (691/100,000 live births) above the national average of 495 deaths for every 100,000 live births, with most of these deaths being attributed to home deliveries and low utilization of skilled maternal healthcare services (Statistics, 2013). Previous studies done in Siaya County to establish factors affecting utilization of antenatal care services associated low uptake of antenatal 756 care with uncertainty about pregnancy, poor communication between health workers and women, cultural and religious beliefs, vague knowledge and quality of ANC services among others (M. A. Pell C, Were F, Afrah NA, Chatio S, et al., 2013; UNICEF, 2013). However, the researcher could not trace any study related to uptake of skilled delivery and postnatal care services in Siaya County despite the worrying trend.
In the quest to increase utilization of skilled maternal health services in Siaya County, the health department managed to scale up implementations of key interventions such as provision of free maternity services, training of healthcare providers on focused antenatal care, 100% scale up of community healthcare services and payment of monthly stipend to all community health volunteers to motivate them to further intensify demand creation (Siaya, 2013;www.hiskenya.org, 2013/14). However despite the investments, utilization of skilled maternal healthcare has remained low; 4 th antenatal care-45%, skilled delivery -52% and postnatal care-30% (www.hiskenya.org, 2013/14) while mothers continue to die from preventable conditions. Taken together the above observation, the aim of this study was to develop and evaluate the effectiveness of context specific interventions to enhance maternal healthcare continuum and optimize utilization of skilled maternal healthcare services in Siaya County, Methods:-Study design:-This was a longitudinal retrospectiveprospective study conducted in 8 selected community units (CU) and their link health facilities in Siaya County. The baseline took 21/2 months, framework development 2 weeks and framework piloting/ intervention took 12 months from 27 th March 2015 to 27 th February 2016. The community health volunteers (CHVs) and the healthcare providers were sensitized on the study before commencement of the study and at every stage updated on the progress and their roles. The mothers were followed up by community health volunteers (CHVs) to ensure that they received all the recommended skilled maternal healthcare services which include 1 st to 4 th antenatal care visits, skilled delivery services and postnatal care services.

Study setting:-
Siaya County is one of the counties in Kenya located in the western region of the Country. It has six sub counties namely Ugenya, Ugunja, Gem, Siaya Alego, Rarieda and Bondo with an estimated population of 935,555 (KNBS, 2009). The county has sex-balanced ratio of female: male (52:48) with the population being predominantly youthful and 23 percent of the population is made up of women of reproductive age 205,356 . The estimated no of pregnant women is 37,592 and estimated deliveries is 37,592(www.hiskenya.org, 2013/14). The study was conducted in selected community Units and their link health facilities. The County has a total of 156 health facilities out of which 122 are government facilities. The County has 187 community units (an equivalent of a sub location) linked to the government health facilities and 2148 Community health volunteers (CHVs) who provide community based healthcare. Each community unit has approximately 10 CHVs who manage about 100 households each. The study sites included Ndori community unit linked to Akala Health Center in Gem Sub County; Nyawita Community Unit linked to Nyawita dispensary in Bondo sub county; West Katweng'a Community Unit linked to Chianda Dispensary in Rarieda Sub County, Gongo community unit linked to Gongo Dispensary in Gem Sub county and Ulafu and Umala community unit linked to Umala dispensary in Alego Usonga sub county, Marenyo community Unit linked to Midhine dispensary and Yiro west community unit linked to Sikalame dispensary in Ugunja sub county. The term community unit referred to in this study is an equivalent of a sub location (MOH, 2010)

Study participants and recruitment:-
A total of 517 pregnant mothers living within the selected community units were purposively recruited for the study. For purposes of knowing the previous maternal health services utilization status, only pregnant mothers who had had experience of giving birth previously (Not pregnant for the first time) were given the chance to participate. The participants were recruited through a combination of random and purposive sampling methodology. Some were invited to participate in the interview after being attended to at the link health facility; others were identified through snowballing technique and others through household's visitation. The sample size was increased by 20% to take care on non-response and fall outs. A final minimum sample size of 480 was obtained. However since the study was longitudinal with the likelihood of having fall out, all pregnant women who were willing to participate were enrolled for the study thus bringing the number to 517.

Study sites:-
The study sites were selected using multistage simple random sampling methodology. From a sampling frame of all community units per sub-county, one community unit was randomly picked from the list in each sub county and having calculated that each community unit would have approximately 50 pregnant mothers, for the study to achieve the recommended target of 400, an additional of 2 community units had to be selected randomly to be included. In total 517 sampled pregnant mothers who were willing to participate in the study were interviewed using structured questionnaires. All the questions were directed to the women's maternal health utilization experience in the previous pregnancy.

Data Collection;-
Data was collected using structured questionnaires focusing on the use of maternal healthcare services (antenatal care, skilled attendance at birth and postnatal care), participant's awareness and knowledge on maternal health, and continuum of care situation.

Informed consent process:-
Detailed information about the study was given to the participants by research assistants before being asked to decide on whether to participate or not. The consent was then read to the participant in the local language (Luo) after which those who accepted to participate were requested to sign. The study questionnaires and key informant interviews were administered thereafter.

Data analysis:-
The quantitative data was analyzed using both descriptive and inferential statistics. The descriptive statistics was used to describe and summarize the data inform of tables, frequencies and percentages. The inferential statistics was used to help make inferences and draw conclusions. Statistical tests including chi-square test, bivariate and multivariate logistic regression analysis was carried out to identify barriers associated with utilization of maternal health care services. All tests of significance was computed at α = 0.05. The Statistical Package for Social Sciences (SPSS) version 22 was used to analyze the data.

Ethical consideration:-
The study was approved by the ethical review board of University of East Africa Baraton (REC: UEAB/05/02/2015).All the study participants parents and legal guardians gave their written informed consents.    In addition, all the mothers who participated in this study had no book/document in their household where health services provided by the CHVs or other providers who visit the household were recorded or referred to for continuum of care purposes. In addition, 434 (83.9%) of the mothers had the opportunity of being visited in their households by other health providers other than the CHV. However, only 6.4% received pregnancy related services from these providers. The strategies outlined in the framework below were implemented for a period of twelve months with intensive follow up after which an evaluation was carried out and results shown below.

Phase 2:-Developed context specific framework;-
The development of the piloted integrated framework was informed by both the baseline findings and the suggestions from the users and maternal healthcare providers. The proposed strategies for increasing utilization of skilled maternal healthcare services were fed into the design of the framework. The framework took into considerations those interventions that target both the supply and demand side barriers to utilization of skilled maternal healthcare services. Included are also suggestions made by key informants and focus group discussants within the selected study sites. The framework contains six prioritized interventions which are interconnected. In line with the available structures, six strategic areas for enhancing maternal healthcare continuum and optimizing utilization of skilled maternal healthcare services were identified by the study and evaluated. They include 1) use of maternal health job aids for guidance at the health facility and community, 2) mother baby booklet, 3) use of referral tools, 4) empowering CHVs to educate the women and provide basic community maternal healthcare services 5) testing for pregnancy at the community/household level and 6) availing documentation tool at the households for reference and care continuum.
The framework explicitly defined how one intervention informs the other and the feedback mechanism. The interventions were addressing both community and facility bottlenecks which included: Increasing women's awareness on importance of maternal health; Availing 24 skilled delivery services at the health facilities; Improving the quality of interaction between healthcare providers and community; Strengthening communityfacility linkage by use of a household log book to document health services provided ; Empowering community health volunteers to educate women and provide community maternal healthcare services and pregnancy testing services at the household to enhance early identification of pregnant women for early initiation of antenatal care, and use of antenatal and postnatal care job aids both the health facilities and community health workers as shown in ii) Baseline findings also revealed that 83.9% of pregnant women had had the opportunity of being visited at their households by other health providers other than the CHVs to offer other services not pregnancy related. This is an indication that opportunities exist that could be used to disseminate maternal health information by integrating the services during the household visits.
i)Every opportunity that either a skilled provider or a community health worker interact with the women should be used for education and review of women's understanding of maternal health. ii) Women should continuously be given information on maternal health and treated with respect so as to reduce fears and increase the uptake of skilled maternal healthcare services. (iii) Every health provider who visits a household having a woman of reproductive health is encouraged to provide information on importance of skilled maternal health services.
(2) Improving quality of i) Cases of missed opportunities exist as evident (i) Previous studies in Siaya County revealed that quality (i) Train providers on respectful maternity care 762 providers and users interaction and positive attitude change in the DHIS, 2004 data that shows that whereas 82% of women attend first antenatal care visits, only 45% attend 4 th antenatal care visits and 52% receive skilled delivery services.
of interaction between provider and pregnant women was poor as women had vague information about services offered even after interacting with skilled providers (Pell C, 2011) (ii) Baseline findings revealed that although 84.9% of the women received antenatal care, only 55.7% and 23.6% received skilled delivery and postnatal care respectively.
(ii) Promote use of antenatal and post natal job aids to enhance quality interaction. iv) Efforts should be made to enhance maternal healthcare continuum by strengthening community-facility linkage by ensuring there is effective communication between the skilled provider and the community health volunteer.( Documentation in the mother baby book to be strengthened as part of communication enhancement).
(3) Introducing the use of a household log book for documentation of all health services provided at the household level (Strengthening communityfacility linkage) (i) There has not been any tool at the household level where services provided at the household are documented thus different providers who visit the household have no reference document for enhancing care continuum and also ascertaining which services have been offered at the household has been a challenge.
i) Reports indicate that all (100%) of households in Siaya County have a CHV who visits on monthly basis and who is paid stipend by the County government. However, baseline findings revealed that 15.3% of the households had never been visited by a CHV which is an indication that some CHVs do not carry out their mandate.
i) Development, sensitization of providers, community and the mothers, and distribution of a household health visitors log book for documentation of all services provided at the household level and for reference. ii) Provide referral tools to all CHVs and provide space for skilled provider's comments and assignment to CHVs in the mother baby booklet. (4) Empowering community health volunteers with knowledge on maternal health and danger signs (i) The majority of the CHVs have not been trained on maternal healthcare and therefore they may not be confident to share the information with the mothers/women during household visits.
(i) Baseline findings revealed maternal health knowledge gap amongst the study participants which may be attributed to lack of opportunities for education on maternal health. The results below indicate that the number who utilized the recommended 4 antenatal care visits or more (≥4) at end line increased by 18%, an indication that the framework was effective in increasing uptake of skilled maternal health services.   The findings on the maternal healthcare continuum situation:-On the continuum of care situation among the mothers, the researcher designed a questionnaire to collect views of the respondents on the same. The mothers' respondents were presented with statements that had continuum of care situation connotations with different answering techniques. The researcher computed percentage frequencies of the responses from the mothers and was tabulated as shown in the table 3 shown. Although it is acknowledged that community-level interventions are potentially effective ways to address the problem at its roots, as decisions to seek and access health care are strongly influenced by the socio-cultural environment (Lawn et al., 2005;John et al., 1998) and that Saving lives depends not only on high coverage but also on the quality of care delivered through the continuum (PNMCH, 2010; Portela et al., 2003), the baseline findings showed that contrary to reports in Siaya County which shows that all households receive CHV services in Siaya County (Siaya County health department performance review report (2014/15), some households (15.3%) had never been visited by a community health volunteer at all. This is an indication of disconnect in the continuum of care (Siaya County health department performance review report (2014/15). However, at end line, the results showed some improvement although 6.5% still did not have an opportunity to be visited by a CHV despite the emphasis that all be followed up. More puzzling was the finding at baseline that, there was no (0%) book/document/evidence in all the households where services provided by the CHVs and other health providers were recorded or could be referred to for continuum of care purposes. However, at end line after intervening, the majority (81.6%) had the household booklets for documentation of all heath services provided at the households. The context specific framework was found to be suitable for increasing utilization of skilled maternal healthcare services. However there is still room to further refine it to include other strategies proposed by previous scholars.
The implications of these findings are that each strategy contributes some percentage increase in utilization of skilled, maternal health care services. Therefore integration of several strategies brings synergy and yields better results. The context specific framework should be considered for programming towards increasing utilization of skilled maternal healthcare services.

Conclusion:-
Based on the findings of the longitudinal study carried out between March 2015 and December 2015, it is evident that no single intervention is by itself sufficient to increase utilization of skilled maternal healthcare services. Context specific, evidence based framework maximizes the benefits of a range of interventions which complement each other resulting in enhanced continuum of care and increased utilization of skilled maternal healthcare services. Declarations