An Investigation into the Implementation Processes of Ghana’s Primary Health Care Strategy: The Community-Based Health Planning and Services Programme.

Background: Ghana has adopted the community-based health planning and services (CHPS) programme as the public health strategy for meeting its universal health goals. The programme is facing implementation challenges that are affecting its expansion within the communities. This research was undertaken to examine the implementation processes of CHPS and suggest solutions to improve its scaling-up within the communities. Method : An exploratory research design was used with a mixed method approach that involved the testing of a hypothesis. Results : The study found that in places with on-going CHPS programmes, there is statistically significant (p<0.001) evidence that the implementation processes are not flawed. However, the district assemblies were selective in the allocation of CHPS zones within the communities. Conclusion : Chieftaincy conflicts within the communities are impeding the community entry aspect of the CHPS policy implementation processes and needs to be addressed by policy makers in the relevant government agencies. Municipalities,


Introduction
The World Health Report of 2013 defined universal health systems along three The community-based health planning and services (CHPS) concept was introduced into the Ghanaian healthcare delivery system with the objectives of improving equity and access to basic healthcare services; improving efficiency and responsiveness to client needs and developing effective intersectoral collaboration on health issues at the district level ). The CHPS programme was seen as a means of improving health service delivery as well as engaging and sustaining the interest of community members in taking charge of their healthcare needs. Eight years after its launching, the programme had been implemented in all of Ghana's districts. However, by the year 2009, it had become evident that implementation challenges were affecting the scaling-up of the programme, and CHPS, as it was originally envisioned, was reaching only 12% of Ghana's households (Binka et al. 2009). In almost all parts of the country, expansion of the programme had either stalled or was incomplete. The If successfully implemented, the CHPS programme would help address a lot of the healthcare delivery challenges Ghana has faced since independence. The country lacks the requisite number of trained medical doctors that are needed for the healthcare delivery system, which often result in delays in seeking care. Also, large volumes of patient load often result in staff fatigue and low morale, which is associated with poor staff attitude, patience dissatisfaction and cases of medical negligence (Pienaar 2016).
Studies have shown that people often migrate from the rural areas to the cities in search of medical care, since most of the clinical healthcare facilities are located in the larger towns and cities (McCormack et al. 2017). A well-functioning CHPS programme would, therefore, not only save the country money, but will also stem the perineal ruralurban migration in the country.
For Ghana to achieve universal health coverage, it is imperative that the CHPS programme becomes successful. Evaluating such success would depend on the availability of empirical data that supports the anecdotal evidence of the achievements of the CHPS programme. The data from this study exposed the strengths and weaknesses of the implementation processes, which would help the district managers recognize areas that need improvement. Policy makers, especially those in the health and allied ministries, are expected to use the data from this study to formulate better CHPS and other health implementation policies for the country. Furthermore, the study is expected to help the Ghana Health Service (GHS) to fashion out better healthcare delivery services to the rural and the peri-urban communities. Finally, this study opens up new research areas in the CHPS concept of healthcare delivery in Ghana and elsewhere. This study examined the implementation processes of the CHPS programme by the Ga East and Ga West municipal assemblies with the objective of identifying the constraining factors affecting its expansion within the communities.

Ethics statement
After satisfying the protocol requirements for the study, ethical approval was obtained from the Ghana Health Service Ethics Committee. Written informed consent was sought from all respondents before taking part in the study and also before interviews were recorded. No oral consent was accepted for this study. Personal identifiers and locator information were not collected and any identifying information accidentally mentioned was removed from the text prior to data analysis.

Research design
The study examined a hypothesis, based on Binka et al, that the CHPS policy implementation process was flawed. An exploratory study was designed for this research with a mixed methods approach involving qualitative and quantitative elements. The quantitative method was chosen because it was deemed useful in providing an objective assessment of the CHPS programme and also to add to the literature on CHPS.

The study area
Two districts within the Greater Accra region were chosen for the study: the Ga West and the Ga East municipal assemblies. The two districts were selected for this study because they are among the oldest local government units in the Greater Accra Region and are therefore endowed with well-established district structures. As well as having functioning CHPS programmes, the two also have long-term experience in district health administration. Furthermore, the selected districts have both rural and urban features, which allowed the study to benefit from an enriched understanding of the interactive processes that evolve between decentralized structures and communities in both the rural and urban settings with respect to health care delivery.
The Ga West municipal assembly is peri-urban in nature, occupies a land surface area

Sampling technique and data collection procedures
With a study population of 640, the sample size was determined to be 240 at 95% confidence level using a digital sample size calculator, the Creative Research Systems Survey software [Version 11]. The study employed two sampling techniques. A purposive sampling technique was first used to select key informants who are knowledgeable about the history and the implementation processes of the CHPS programme. A simple random sampling method was then used to select respondents from among the purposefully selected participants. Three instruments were used for data collection: a semi-structured questionnaire, in-depth interviews and focus-group discussions.
A six-point Likert-scale type questionnaire ranging from (1=strongly agree to 6=strongly disagree) was administered to the selected participants. Two hundred and forty (240) such questionnaires were distributed among the study participants out of which 196 (81.7%) were returned. Following this, five (5) individuals from each of the municipalities were randomly selected from among those who responded to the questionnaire for in-depth interviews (IDIs). Seven (7) randomly selected individuals from within the study participants of each of the districts were used for a focus-group discussion (FGD). All qualitative data were audiotaped using digital audio-recorders.

Analytical procedures
The study examined the hypothesis that the CHPS policy implementation processes were flawed by using five variables: the extent of CHPS marketing in the communities; fidelity of the implementation processes to the 15 policy steps; roles played by the traditional authorities (TAs) in the implementation processes; the extent of community participation in the implementation processes and the district assemblies' (DAs) management of the implementation processes.
The main emerging themes in all the transcriptions from the recorded interviews were identified and coded into "NVivo 10" software package for qualitative data analysis.
The study explored the distribution of socio-demographic/economic factors using frequencies and percent-frequencies. Bivariate analysis, based on Chi-square test of independence and the Fishers Exact test, were conducted to determine the association between each of the socio-demographic factors and the prevailing condition of the CHPS policy implementation processes. The null hypothesis was tested using onesample t-test after the distribution has satisfied the normality test. All the statistical analyses were performed using Stata version 14 (StataCorp, College Station; Texas, USA) and p-value less than 0.05 was considered statistically significant. The quantitative data was presented in the form of tables (frequency tables) and charts for easy interpretation and analysis.

Marketing of CHPS
On the question of how well the CHPS programme was marketed in the communities, the general consensus was that the programme was not well advertised. Majority of the respondents were of the view that previous marketing of the programme was better compared to the present.

Fidelity to the CHPS policy implementation steps
Respondents were of the view that the district assembly tried, but failed to follow the 15-implementation steps in the CHPS policy document. They were unanimous that the implementation steps were cumbersome and time-consuming. Some respondents also felt the implementation steps were not well-aligned and sometimes created confusion during the implementation process.

Role played by the traditional authorities (TAs)
There were mixed responses to the roles played by the traditional authorities in the communities with respect to the implementation process. Whereas some respondents thought that the chiefs, in particular, did not play any meaningful role with the process, others were of the strong view that the opinion leaders played key roles. Some contended that the chiefs did not want to be seen as hijacking the process, hence their support was minimised in the public view.
"The chiefs were supportive of the programme, but they didn't want to be seen as the ones in charge so they provide support on the quiet. They are the ones who summon the people for durbars on CHPS activities. Some even provided office space and accommodation for CHOs in the CHPS zones." (Opinion

Community participation
Respondents agreed that the community members played a satisfactory role in the implementation process, especially in the rural areas. Some, however, were of the view that community participation in the urban and peri-urban areas was not encouraging.
"The community members offered their skills and labour to support the CHPS programme. This was especially so in the rural communities. Support from the urban and peri-urban areas was challenging as the people complained of not having time for CHPS activities." (CHMC, FGD)

District assembly management of the implementation process
There was overwhelming agreement that the district assemblies (DAs) managed the implementation processes quite well. There were, however, a few dissenting views from respondents who thought the assembly could have done better by involving other stakeholders within the communities.

Socio-demographic characteristics of study participants
One hundred and ninety-six participants took part in the study with the Ga West providing 102 (52%) and Ga East contributing 94 (48%). Female participants constituted 124 (63%) with males forming 72 (37%). The participants aged between 28-and-60 years and most of them (58%) were married with 74% having at least one child. The educational level of the participants was high with 95% having had formal education out of which about half (50.5%) have had tertiary education. About a quarter of the participants (28%) were self-employed and their income levels ranged between GHC 600 ($115) and GHC 1,500 ($290) per month (Table 1).

Responses on the assessment of the CHPS programme
On the question of the marketing of the CHPS programme within the communities, majority (79.6%) of the respondents were of the opinion that the programme was not well-marketed. As to how well the implementation process focused on the milestones of the CHPS policy document, 62.2% of the respondents were of the view that the process did not follow the CHPS milestones. A significant minority of 37.8%, however, disagreed. Only 41.8% of the respondents were of the opinion that the traditional authorities played any major role in the implementation processes. Majority (58.2%) of the respondents were of the view that the traditional authorities did not play any major role in the implementation processes. However, majority of the respondents (61.2%) were of the view that community participation in the implementation process was low.
Nevertheless, there was overwhelming support for the district assemblies' management of the CHPS programme implementation processes with an 89.8% approval rating. The frequency distributions of the responses are highlighted in Table 2 and graphically presented in Figure 2.

Statistical analysis of the quantitative data
Statistical analysis of the quantitative data indicated that the CHPS policy implementation processes were not flawed, as indicated by 113 (57.6%) of the study participants (Table 3).

Table 3: Evaluation of the CHPS programme in Ga East and Ga West Municipal
Assemblies.

CHPS POLICY IMPLEMENTATION
Not flawed 113 57.65

Results of hypothesis testing
Based on the one sample t-test and the corresponding p-value, this study rejected the null hypothesis that the CHPS policy implementation processes were flawed (p<0.001).
The results showed statistical evidence that the CHPS policy implementation processes were not flawed (Table 4). : Null hypothesis of population mean score pegged at the 25 th percentile value indicating that the implementation process was flawed.

Association between the socio-demographic characteristics and the flawlessness or otherwise of the CHPS policy implementation processes.
Bivariate analysis using Fisher's exact test indicated that married people, those with high levels of education and people with children believe that the CHPS policy implementation processes were not flawed. On the other hand, the single, the selfemployed, the less educated and low-income earners were of the opinion that the processes were flawed. The results further showed that females and people less than 40 years old (the youth) were less likely to see the processes as flawed compared to males and the elderly. The analysis also revealed that the most significant demographic characteristic associated with the flawlessness or otherwise of the CHPS policy implementation process is the educational level of the individual (Table 5).

DISCUSSION
The study found enough statistical evidence to conclude that the CHPS policy implementation processes were not flawed (p<0.001). However, the process faces challenges that are making it less effective than expected. The CHPS implementation process centers on community participation, with the chiefs and the traditional leaders of the communities playing a critical role to its success. It also requires the ready availability of information to the stakeholders, especially the community members. The study found that the implementation process was not well-marketed in the communities There are defined roles for the chiefs in the policy implementation process in which they are to organise durbars and select community health volunteers (CHVs) and community health management committee (CHMC) members. It appears that the chiefs and the traditional leaders, aside selecting the health volunteers and the health committee members, do not play any active role in the implementation process. This could mean that the traditional authorities lack the understanding that for the programme to be sustained, they need to put in place measures to ensure that the selected volunteers performed their duties.
The study revealed that the traditional authorities expected to be given assigned roles by the district assembly leadership in order to perform other roles in the programme because they did not want to be seen as usurping the powers of the state. The nonspecific involvement of the traditional authorities could be due to the perennial conflict that exists between traditional authorities and state functionaries (Inkoom et al. 2004).
Since the advent of the

Conclusions and recommendations
In this study, it has been shown statistically that the CHPS policy implementation processes were not flawed. However, the main reason for the inability of the programme to be scaled-up is the numerous chieftaincy disputes within the communities, which are hampering the community entry processes of the policy. The study also found that fidelity to the CHPS policy implementation steps is challenging because the steps are cumbersome and should be simplified. Furthermore, the study revealed that the implementation of the CHPS programme could be successfully carried out by the inclusion of other leadership groups in the community entry process beside the chiefs.  Responses to the CHPS policy implementation processes