An innovation to improve health outcomes in Amajuba district, KwaZulu-Natal, South Africa

ABSTRACT

Systems’ thinking places high value on understanding the context. This study focused on the collection of disaggregated data in order to understand the context, to facilitate improvement of health outcomes. The aim of this article was to assess the implementation of municipal ward-based health data collection (disaggregated data) and health care workers’ perceptions of this data collection process. This cross-sectional study used mixed methods in Amajuba district. The participants were professional nurses at the Primary Health Care level. Of the 131 respondents, 123 (93.9%) collected municipal ward-based health data, and found it useful. Opportunities for improving data collection were identified. Disaggregation of the data at ward level contributes to a better understanding of the target population’s health, assists planning for health needs and enables provision of targeted interventions in order to improve health outcomes, to prevent financial regression and waste of health resources.

KEYWORDS:

• Disaggregated data
• National Health Insurance
• financial risk protection

1. Introduction

South Africa is planning to implement National Health Insurance (NHI). The first five years of piloting NHI are for testing innovations, investigating what works, for whom and what can be scaled up. Amajuba district in KwaZulu-Natal is one of the 11 pilot sites. In view of the health transformation in South Africa which aims to provide universal access to health services, this warrants a concerted effort and bold interventions to mitigate the challenges of the rising cost of health services. Inequities exist with regards to access to health services, with the uninsured population where the majority having limited access (Department of Health, 2015).

NHI is a funding mechanism to improve health outcomes (Department of Health, 2015). The main deliverable function of NHI is universal access to health services irrespective of whether a person is insured or uninsured. Universal access can be achieved through increasing access to health services and prevention of financial regression by families. Primary Health Care (PHC) re-engineering is the cornerstone of NHI (emphasising prevention of disease and promotion of health). Its ultimate aim is to create a health system with a shift to cost-effective health care, focusing on disease prevention, health promotion and rehabilitation that is closer to communities and households (English et al., 2011). Furthermore, it promotes increased access to health services through the following care of health professionals, namely school health teams, ward-based outreach teams, district clinical specialist teams, contracted general practitioners and pharmacy assistants (Department of Health, 2011). The NHI pilot period is 14 years, and the first five years of NHI were aimed at testing innovation and assessing what works, for whom, under what circumstances and what can be scalable. Hence, the innovation of collection of the municipal ward-based health data in Amajuba district in order to understand the local context and improve health outcomes.

The increase in the life expectancy and the unstable economic climate pose a further burden and potential financial risk to the health system. Hence, the need for an innovation to collect municipal ward-based health data which can increase responsiveness to community needs whilst managing scarce resources, through targeted interventions, in order to improve health outcomes. The increase in the disease burden of infectious and chronic diseases (Department of Health, 2011) and South Africa’s failure to meet the Millennium Development Goals also necessitated this innovation to mitigate current challenges facing the health system. It is envisaged that this innovation will improve health outcomes, thereby contributing to the achievement of the Sustainable Development Goals 2016–30.

An effective health system is responsive to the community’s need as people are central to the health system, both as mediators and as beneficiaries (De Savigny & Adam, 2009). Strengthening health systems is thus increasingly on the political agenda worldwide since it is viewed as a panacea for improving health outcomes (De Savigny & Adam, 2009; Nutley, 2012). Despite this perspective on the need to improve health systems, many health managers lack the capacity to measure or understand their own weaknesses or constraints, and to elicit which activities are directly related to improved health system performance, improved health outcomes and sustained impact in different settings (Travis et al., 2004; WHO, 2007; De Savigny & Adam, 2009; US GHI, 2011). This effectively leaves policy-makers without sufficiently sound information as to what they can or should actually strengthen (De Savigny & Adam, 2009). In addition, good planning and decision-making require a comprehensive health information system, which captures relevant data and produces timely useful reports in order to improve the health system and to improve health (De Savigny & Adam, 2009).

In South Africa, provinces are divided into districts and sub-districts or local municipalities which are further subdivided into wards. Amajuba district comprises three sub-districts and 46 wards. Municipal ward-based health data collection is an innovation implemented in Amajuba district which is aimed at improving health outcomes. There are inequities and scarcity of resources in Amajuba district, which this innovation seeks to overcome. The municipal ward-based health data collection requires collection of data and documenting the municipal ward so as to determine the prevalence of different health conditions in each ward. The aim is to disaggregate the data in order to understand the context and increase the responsiveness to actual community needs. Health care workers are expected to document the municipal ward of every patient to whom they attend. This requires collecting the patient’s data, collating, analysing and converting this information for decision-making in the district. In each PHC facility consultation room there is a list of ward numbers and the names of the Ward Councillors, should a patient not know the name of the municipal ward but knows the Ward Councillor. This assists to ensure complete records. Every month the data are collated and the wards from where the patients with various health problems are coming are assessed. A summary sheet is completed monthly on the disease burden of the facility’s catchment area. Should they identify an infectious disease then they need to promptly report – for example, a child with measles – in order to immediately screen, notify and treat the contacts. The facility reports are submitted to the district health office which is required to give quarterly feedback to the PHC facilities unless there is a population health threat which needs immediate action.

Each facility develops a programme of action to address the health needs. They implement activities and document them as a record of targeted interventions. The health transformation strategy in South Africa emphasises PHC re-engineering, whereby the recently instituted Ward Based Outreach Teams (WBOTs) play a pivotal role in taking services to the community to improve access to health services and to prevent financial regression by families (Department of Health, 2011). These teams comprise a professional nurse and at least two other health workers. The emphasis is on health promotion and disease prevention, since, for example, no child should die of malnutrition or preventable diseases. The information from the WBOT when shared with the PHC facility during information committee meetings can enhance understanding of the context and facilitate facility responsiveness to the community needs.

We used the theoretical framework adapted from Lippeveld & Sauerborn (2000) which entails data collection, processing (analysis), presentation (report), use (decision based on information) and feedback to the data collectors. The information cycle systematically describes how data are handled and applied in each stage of the cycle to ensure timely generation of useful information for evidence-informed decisions. This model was selected because it addressed two challenges experienced in Amajuba district, whereby data were consistently collected but not used for action to improve health outcomes. Moreover, data analysis is essential to ensure quality data inform decisions, as actions can be off track and not address the needs if data are inaccurate. Furthermore, this principle is applicable at all levels of patient care, community, health facility and district levels.

This article describes the implementation of municipal ward-based health data collection and the facilitating factors and barriers to the provision of useful information that can improve the health system’s ability to contribute to improved health outcomes. The aim of this article is to assess the implementation of municipal ward-based health data collection and health care workers’ perceptions of this data collection process.

2. Methods

The study was conducted in Amajuba district, one of the 11 districts in KwaZulu-Natal, South Africa, which comprises three sub-districts or local municipalities, namely Dannhauser, Emadlangeni and Newcastle.

The 46 wards in the three sub-districts are a hybrid of rural, urban and peri-urban areas. Dannhauser is rural with one ward out of the 11 which is peri-urban. Emadlangeni is rural with farms and one urban ward out of the four. It is sparsely populated and larger in size than the other two sub-districts combined. Newcastle has 31 wards of which 23 are urban, six are peri-urban and two are rural.

This was a cross-sectional study using an explanatory sequential mixed-methods approach, collecting both quantitative and qualitative data from the professional nurses providing PHC services. The study population was from 25 fixed PHC and seven mobile clinics and PHC services provided by nine WBOTs. It comprised 24 operational managers who are professional nurses, 98 professional nurses, nine professional nurses who were WBOT team leaders and four PHC supervisors working in public PHC service delivery sites. Purposive sampling was used for the focus group discussions (FGDs).

A random sample of 131 individuals was selected for the quantitative study from the total of 198 professional nurses. This sample size was determined using a two-sided 95% confidence interval with a precision of 5% when the actual proportion is near 50% which assumes maximum variability (Machin et al., 2009).

A self-reported semi-structured anonymous questionnaire was used to collect the quantitative data, which investigated the implementation of municipal ward-based health data collection comprising: collection, analysis, reporting, usefulness and feedback received. All quantitative data were coded and captured in EpiData Version 7 and analysed using STATA Version 13.

Five FGDs were conducted: PHC supervisors (one FGD of the total four individuals), operational managers (two FGDs of five persons each) and two FGDs of all the WBOT leaders (five and four individuals in each group). The total number of participants for the FGDs was 23, comprising 22 females and one male who was a WBOT leader. The trained moderator of the discussions used an interview guide to cover the main themes and, with the permission of the respondents, the discussions were audio recorded. Subsequently, the process entailed transcribing of the audio recordings and reading, re-reading and analysing the data. Thematic analysis was used in order to code and understand the concepts.

The research assistants were trained on municipal ward-based health data collection, and on the quantitative instrument. Further training was undertaken on conducting FGDs and the use of the guiding themes, which included how to pursue a theme until saturation whilst ensuring participation of all the participants. In addition, a document review was done using each of the PHC Registers (facility, mobile and WBOT) and summary sheets from January to December 2015, to assess whether and the extent to which municipal ward-based health data collection is implemented. Triangulation of the data was undertaken using the data from the questionnaire, FGDs and the document review to elicit similarities and differences.

The University of KwaZulu-Natal Humanities and Social Sciences’ Research Ethics’ Committee provided ethical approval (HSS/0676/015D). The main ethical consideration in this study was confidentiality of the information obtained from the respondents. Informed written consent was obtained from all of the study participants. Written permission to access data and the health facilities was granted by the Epidemiology, Health Research and Knowledge Management Unit, KwaZulu-Natal Department of Health and Amajuba District Department of Health.

3. Results

3.1. Quantitative data

The majority of the professional nurses (123/131, 93.9%) collected data but very few (n = 9, 7.8%) analysed this weekly, with about half (63/115, 54.5%) doing so monthly and 30 individuals (26.1%) only as the need arises. The majority (110, 84%) submitted this data monthly to the district but only about a third (43, 32.8%) received feedback from the district, as illustrated in Table 1.

Table 1. Professional nurses’ responses regarding municipal ward-based health data.

	Frequency, n (%)
Collection of municipal ward-based health data, n = 131
Yes	123 (93.9%)
No	7 (5.3%)
Do not know	1 (0.8%)
Frequency of analysing municipal ward health data, n = 115
Weekly	9 (7.8%)
Monthly	63 (54.8%)
Quarterly	13 (11.3%)
As the need arises	30 (26.1%)
Rationale for analysis of municipal ward-based health data, n = 116
Prevalence	38 (32.7%)
Trends	27 (23.2%)
Health-seeking behaviour	14 (12, 1%)
Prevalence + trends	14 (12, 1%)
Prevalence + trends + health-seeking behaviour	23 (19.8%)
Frequency of submission of municipal ward-based health data to district, n = 131
Monthly	110 (84%)
Quarterly	11(8.4%)
Sometimes	4 (3.1%)
Never	6 (4.6%)
District’s responsiveness regarding feedback to professional nurses, n = 131
Yes	43 (32.8%)
No	58 (44.3%)
Sometimes	20 (15.3%)
Do not know	10 (7.6%)

3.2. Findings from the qualitative data

3.2.1. Purpose of municipal ward-based health data collection

The respondents understood the rationale for the implementation of data collection at the ward level. They explained that this would assist them to understand the disease burden and disease trends per municipal ward. Some of the respondents also emphasised the importance of analysing the data in order to identify which interventions were needed in order to provide such in a timely manner. Many of the respondents were aware of the importance of knowledge of the local epidemiology in order to address the burden of disease and to promote the community’s involvement. The majority of the respondents thus understood the purpose of the collection of data at ward level so as to prevent diseases from spreading and to promote health. Table 2 presents some of the comments from the participants.

Table 2. Participants’ comments on ward-based data collection by job category.

Job category	Comments
Operational managers	To understand the disease burden and the disease trend per municipal ward, we need to understand the disease profile and to determine which age group is affected and why, and involve the community to reduce the burden of disease. It was important to understand the epidemiology and demographic profile of the population served in order to mitigate the disease burden To identify the disease burden of the specific ward and to have measures of how these problems or diseases are going to be addressed to the community that we are serving You will be able to do the budget according to the profiling of the clients. If you’ve got the number of the clients you are able to budget for that particular disease when planning
WBOTs	To identify problems experienced by the community focusing on the ward, for example if they do not immunise their children because it causes diseases To give us an alarm for which places are vulnerable, need attention; what do we need to do? To understand the clinic catchment, how many people the clinic sees? Identify diseases on the peak The purpose is to identify problems of the wards and plan for them and for relevant interventions. For example, if you identify a specific ward has TB, you are able to plan awareness to prevent the spread of TB and to plan for health promotion. So as ward-based teams we know exactly which are the main problems facing which ward It helps the district plan according to community needs As outreach teams it will also help us to trace specific clients who are supposed to be seen at the clinic, it facilitates location of the client
PHC supervisors	Purpose is to plan for our communities; different wards, when we have identified the need we then plan for them and then intervene, then our people will live long Our strategies are going to be focused because we are going to be focusing on that particular ward and our resources on that particular ward Another reason is for campaign purposes, because if we have identified that there is an outbreak of diarrhoea in a certain ward, so the personnel, Department of Health personnel, will go there to do the campaign

TB = tuberculosis.

3.2.2. Perceptions of current functioning of municipal ward-based health data collection

The general perception was that improvements were required. Moreover, the new tool designed to collect the data was found not to be user friendly. In some wards where it was working, the data were being collected and the health care workers were responding according to the findings and providing health education. Some found it useful, stating that it enabled them to disaggregate the data per municipal ward and plan accordingly. Moreover, they explained that because they had been oriented to the use of the tool, they were able to implement it. Such participants also extended their outreach to the ‘war rooms’, a weekly discussion forum held at ward level for inter-sectoral collaboration, which offer a platform for presenting reports and provision of health education. Other respondents mentioned that the ward-based health data collection was not functioning effectively. Their reasons included the lack of observable results of improvement in health status. They emphasised that in-service education on the use of the new tool was required. There were a range of other problems reported, in that feedback was not received from the district office and cross-border patients from Mpumalanga province could not be followed up.

The WBOTs described how they implemented the data collection and reported that monthly meetings with the clinic’s operational manager were held to discuss the statistics. However, the challenge was that two weeks after submission of their monthly data they are required to resubmit this and they do not get any feedback. Furthermore, the WBOTs were not members of the facility information review committee at each clinic. A problem identified was that some of the work done was not captured as it was not listed in their tool. The non-participation of some WBOTs contributes to loss of valuable data that can enhance understanding of the context and inform decisions to improve health outcomes.

The majority agreed that to improve the effectiveness of services requires WBOTs in every municipal ward, which is not currently occurring. Table 3 presents participants’ comments by job category.

Table 3. Participants’ comments on current functioning of municipal ward-based health data collection.

Job category	Comments
Operational managers	The previous tool was user friendly, it was better, people were able to collect data according to the disease that was prevailing at that particular time It is functional because now we can be able to understand your own wards, how many HIV, TB patients we are having and the ages. Then it is good for you because you can be able to plan maybe for an awareness in case you have them, teenage pregnancy. I think it is functional I have always had a problem with this. Perhaps I am one of the people who need a workshop so that things will be fine. I have identified that you sometimes repeat one and the same person and you will perceive that the number is high and you find out you repeat, one person on the very same visit. Sometimes you would not know that person has come and visited that facility the previous two weeks and in the same month you report that person double
WBOTs	It is functional, at month end we discuss statistics with staff including myself, identify gaps, for example, if vitamin A falls, and discuss outbreak of diseases, trying to intervene It is helping because we are able to work according to what we found. Where there is a need go to the war rooms, increase awareness and talk about the information we want to share It is difficult to know the health outcome, much as you write a referral letter, you do not know whether the patient attended or not As the WBOTs, there are clients that we see like defaulters that are not in our data collection tools. There is nowhere where we record and submit, for example, defaulters. There is no specific tool for these clients. Like, defaulters that we trace are not recorded anywhere; so we cannot use that information There is a communication breakdown because we submit data but after a week or two we are asked to submit the same data. I feel data should be received by the relevant person at the district
PHC supervisors	It is not used as it is supposed to; they do not intervene as they are supposed to. If it were, we would be seeing a lot of awareness, a lot of campaigns or interventions Data are mostly under-reported What I noticed is that our war rooms are not properly represented by different stakeholders. You find the Department of Health as the biggest percentage, the other stakeholders do not attend war rooms, so making it difficult really to discuss the problems re ward-based data

TB = tuberculosis.

3.2.3. Factors which promote or inhibit municipal ward-based health data collection

The FGDs identified that the most important factors which promoted the collection of municipal ward-based health data were the involvement of the community when clients provided accurate addresses, thus permitting follow up, the availability of appropriate data collection tools, provision of in-service training for all staff about the municipal ward-based health data and then conducting regular information review meetings and deciding on campaigns. Other factors, although reported by fewer participants, were the need for a central registration of patients to avoid duplication, proper community profiling and the availability of an electronic patient record. These findings were similar across all of the FGDs.

The most common factors cited by the respondents in the FGDs which inhibited data collection were the lack of understanding by the community as to why municipal ward-based health data were collected, that not all municipal wards have WBOTs, a lack of transport for WBOTs, poor recording, under-reporting and late submission of data.

Patients who do not present a referral letter when referred by the WBOTs were identified as a problem, since these referrals were not captured and this skewed the contribution of the WBOTs in respect of prompt identification of health problems and referrals. Not all data elements pertaining to the work of the WBOT were captured. It was interesting to note that the inhibiting factors relating to the work of the WBOT were only mentioned by the WBOTs themselves. This confirmed that at the facility level they were not working collaboratively as a team, since the PHC supervisors only mentioned that not all wards had WBOTs.

Once the ward-based health data were submitted to the PHC facilities for collation, the data analysis was limited due to the absence of administration clerks in all facilities. Another concern was that operational managers were not validating the data before signing this off and respondents also reported that there was an absence of stationery. These concerns were identified by all of the FGD categories. The lack of analysis deprived them of the opportunity to understand the disease profile per municipal ward, hence they could not respond to the community’s needs. They believed that the municipal ward-based health data collection could improve if there were data capturers in all facilities, no shortage of stationery and if information committee meetings were held monthly at the facility level before the data were signed off.

The majority of participants perceived the collection of the disaggregated data as beneficial for addressing specific community needs in order to improve health outcomes. The participants embraced the understanding of the context as important so that they can make evidence-informed decisions on provision of health services. There was a general perception that WBOTs are instrumental in improving access by provision of targeted interventions for the community in need of health services. Some promoting and inhibiting factors for collection of disaggregated data were identified; participants gave suggestions as to how inhibiting factors can be mitigated, such as participation of WBOTs in the facility information review meetings to ensure completeness of data.

3.3. Record review

3.3.1. Data collection

The research assistants discovered that there was erratic documentation of the patient’s municipal ward in the registers in a third of facilities. Monthly summary sheets of collated data, however, were available in all of the facilities.

3.3.2. Data analysis

The data analysis was not done by all of the facilities, but 24/41 (58.5%) had analysed their data and there was evidence of this data analysis in the PHC clinic information committee review minutes and the communication book.

3.3.3. From the record review

There was evidence that not all of the PHC clinics collected the data. Further, the reporting frequency varied although it was supposed to be monthly. It was noted that the WBOTs reported monthly, unlike the fixed facilities and the mobile teams which had varied frequencies and formats of reporting. For example, all fixed facilities used bar graphs to present child health indicator performance, and dashboard indicators were developed based on poorly performing child indicators and were monitored monthly.

3.3.4. Use of municipal ward-based health data for decision-making

Only 20/41 (48.8%) facilities who implemented municipal ward-based health data collection used the information to decide what to target and which interventions to implement. A record of these targeted interventions which had been implemented was provided. Some facilities used the facility information review committee book to document this, whereas others captured the information in the communication book.

3.3.5. Feedback to data collectors

In about a third of facilities there was evidence of feedback from the district. Lack of feedback deprived the facilities of understanding the disease burden of the municipal wards as patients may and do access health services outside their catchment area.

4. Discussion

The collection of disaggregated municipal ward-based health data in Amajuba district is an innovative approach which aims to contribute to evidence-informed decisions whereby targeted interventions can be executed to improve health outcomes. It differs from the District Health Information System (DHIS) which is designed to collect aggregated data, since the DHIS only disaggregates data to a facility level. This study found that such data collection of disaggregated ward-based health data is feasible, and useful.

4.1. Positive findings

The majority of the participants understood the rationale for collecting municipal ward-based health data. They found it beneficial in their practice and explained that disaggregated data enable them to understand the context, disease burden and trends. This notion is supported in the literature, in that information is relevant only if it is used to resolve a local problem or it assists in generating innovations that resolve a local problem (Abouzahr & Boerma, 2005; Stansfield et al., 2006). Moreover, understanding the health systems’ contexts is a necessary prerequisite to scaling up new programmes and achieving sustained success (Atun & Menabde, 2008).

Many participants stated the importance of understanding the epidemiology and demography of the ward in order to plan targeted interventions. Some respondents stated that the collection of municipal ward-based health data can influence resource allocation.

The participants emphasised the importance of community involvement when dealing with health problems, and this is confirmed in the literature as a principle of PHC (De Savigny & Adam, 2009; Department of Health, 2011). The policy of PHC re-engineering is an enabler to achieve the aim of NHI, which is universal access (Department of Health, 2011). Moreover, the social determinants of health need a collective approach from different service providers in mitigating them. Health promotion and health challenges were discussed by WBOTs and PHC supervisors in the war rooms, which are led by the ward councillor or a nominated government employee.

4.2. Positive factors influencing the collection of municipal ward-based health data

The value of disaggregated data per municipal ward was a common thread reported in the quantitative, qualitative data collection and the record review. Early warning signs can be identified through the data analysis and informed decisions taken to execute targeted interventions, such as timely identification of an outbreak in a specific ward. When analysing these data, health care providers were looking for prevalence, trends and health-seeking behaviour and this enabled those who utilised the data to provide targeted interventions, through their understanding of the context and in planning for the wards’ health needs. The FGD participants wanted to know about the context, the myths and the disease burden and to have a comprehensive picture of the population they served, and they emphasised the importance of community participation.

Community participation plays a pivotal role in encouraging communities to take responsibility for their own health, which can contribute to improving health outcomes. The FGDs identified that the most common factors which assisted the collection of municipal ward-based health data were community involvement, with clients providing accurate residential addresses. The emphasis of the data collection was on health promotion and disease prevention. The minutes of the war room meetings showed that these regular meetings of service providers and other stakeholders provided an opportunity and a platform for discussing health needs.

All of the participants valued the contribution of the WBOTs in increasing access to health services and provision of targeted interventions. Hence, they recognised that there is a need for all wards to have WBOT coverage in order to increase responsiveness to community needs. This is supported in the policy paper as WBOTs form a pivotal part of South Africa’s PHC re-engineering strategy (Department of Health, 2015). The WBOTs in turn valued their own work and wanted all aspects of their performance to be captured, and requested to participate in the facility information review meetings. Creation of an enabling environment to share data was valued. The study identified this aspect as providing an opportunity to improve service delivery.

The availability of appropriate data collection tools was identified as a positive step but the majority of participants complained about the new monthly summary tool, which needed changes. In-service training for all staff about the municipal ward-based health data has positively influenced the collection of the municipal ward-based health data. Such training needs to be ongoing and to highlight and inform staff of the results of interventions that they have implemented.

4.3. Negative factors influencing the collection of municipal ward-based health data

Although data collection tools were available in all service delivery sites to fulfil the policy imperatives, not all participants collected municipal ward-based health data. The monthly summary tool was a standardised tool populated with DHIS indicators, and it was supposed to identify the disease burden in that catchment area. Data collection is the first level in the information cycle model (Lippeveld & Sauerborn, 2000; Heywood & Rodhe, 2007) and occurs at the point of contact between the clients and care providers, where information about the clients is recorded in the register. The results show that not all health workers collected the municipal ward-based health data, and this was commensurate with the qualitative data and confirmed by the record review which showed some erratic documentation of the municipal ward in some facilities. The inhibiting factors regarding collection of the municipal ward-based health data were clarified in the discussions, including that the tool is not user friendly.

As mentioned previously in some facilities, collation of all data did not take place since the WBOTs did not attend the information review committee meetings and some operational managers signed off data without validating it. Under-reporting is another challenge and some of the work executed by WBOTs was not captured, and this negatively influences the financing of the health facility. When some information is missing the extent of the service provided and the resource requirements cannot be accurately measured, as information can be used to increase overall resources for health (Stansfield et al., 2006).

The study also noted that some participants reported data without analysing it. Yet data analysis is essential to ensure quality data inform decisions as actions can be off track and not address the needs if the data are inaccurate. At a practical level, the lack of resources such as stationery and administration clerks inhibited effective municipal ward-based health data collection. The WHO confirms that good quality, timely and contextual information is needed by all the components of the health system to enable the managers to plan, implement and evaluate programmes (WHO, 2008). The fact that there was erratic reporting from some facilities and variances in the reporting frequency is of concern since just above half of the respondents reported monthly. The document review found evidence for only about a third of all fixed and mobile facilities reporting monthly but it was available for the WBOTs. Despite the directive on the frequency of reporting and the importance of timely, accurate, contextual data, gaps were identified.

The study identified challenges with the dissemination of information. Operational managers were not sharing information with the teams. The report that in some facilities information meetings did not include the WBOTs was confirmed by the document review. This means that they do not know the health challenges in their catchment area, yet collection of municipal ward-based health data was intended to empower WBOTs to understand the disease burden, in order for them to be responsive to the community needs. This emphasises that at the facilities they were not working as a team and that not all employees at PHC delivery sites were aware of the new processes that had been implemented. The need for continuing engagement and dissemination of information was emphasised by the study.

The erratic attendance by other sector departments in the war rooms deprived the health service providers of an opportunity to share child health challenges and mitigation strategies. Yet war rooms are an ideal platform for addressing social ills, taking into consideration the social determinants of health.

It was evident that most employees were deprived of their right to receive timely feedback despite their regular submission of data. These findings were similar in the quantitative, qualitative data collection and the record review. This compromised the understanding of the burden of disease in their wards as patients’ may also access health services in other facilities. This was confirmed during the discussions as one of the inhibiting factors in the implementation of the municipal ward-based health data collection. Feedback from the district office could provide a more comprehensive picture, showing trends and prevalence. Employees have the right to expect timely feedback from the District Facility Information Officer if they have submitted their municipal ward-based health data reports (Heywood & Rodhe, 2007). Further, the health outcomes of cross-border patients from Mpumalanga province could not be elicited as referral forms are written but no feedback was received as this requires cross-border cooperation between the two provinces, KwaZulu-Natal and Mpumalanga.

4.4. Strengthening the health system

The use of explanatory sequential mixed methods and triangulation of data assisted in identifying the positive and negative factors regarding the collection of ward-based data. Health systems’ strengthening requires contextual analysis as the context shapes the responsiveness of the health system. The collection of municipal ward-based health data can enable health care providers to better understand their context, so that they can make evidence-informed decisions as to how to respond to the community needs through targeted interventions. Evidence from sub-Saharan Africa on saving lives with targeted health interventions shows remarkable achievements through facility-based and outreach services (Victora et al., 2003; Friberg et al., 2010). Most respondents perceived WBOTs as instrumental in enhancing the understanding of the context and the provision of targeted interventions, stating that the WBOTs were established to provide direct services to needy households. Prioritisation of interventions towards the needy can enhance success (Friberg et al., 2010). Targeted interventions have been shown to be effective and contribute to public value if they are well designed and implemented to address the specific health challenges (Victora et al., 2003).

The provision of in-service education whereby municipal ward-based health data are a regular agenda item, offering opportunities for discussion, is necessary. This can ensure that quality, contextual and timely data are shared in order to inform decision-making at the local level. Health care providers need to be taught about the health system and how information can influence other components of the health system, such as informing resource requirements and service delivery. In this study some respondents stated that such data inform resource allocation. Further issues such as the provision of stationery can enhance collection of municipal ward-based health data. Respondents emphasised the importance of facility information review committee meetings, and that they need to involve the WBOTs who to date were not participating in some facilities.

5. Limitation

The study was carried out in one district at PHC level, hence generalisations cannot be made to all levels of care or other districts.

6. Recommendations

Efforts to implement municipal ward-based health data collection whilst addressing the need to improve the data collection form need to be scaled up. It is necessary to strengthen management support and commitment, including appropriate resource allocation such as transport and response to data submissions. The focus should be on reviewing the data summary tool, data quality, timeliness and including the allocation of the required resources, to enable disaggregation of data such as human resources for health.

Regular participatory in-service education, tailored to institutionalise collection of municipal ward-based health data, needs to be conducted. Further strengthening relationships between staff members in order to promote team spirit to ensure completeness and ownership of data is imperative. Monthly information review meetings should be strengthened to encourage inclusiveness and in-depth analysis of the data. The district information team should attend these meetings, and prioritise the monitoring of the child health indicators. Monthly reported data must be validated by the operational manager before signing off, and this should be monitored.

In-service training needs to inculcate a culture of information use at the different organisational levels. This must include feedback mechanisms which are essential for the generation of quality data. Data management needs to be a key result area for all data handlers, in order for the health indicators to be accurate and to enable management to monitor whether health is improving.

7. Conclusion

This assessment has documented particular inhibiting factors in collecting municipal ward-based health data in Amajuba district and has identified opportunities for improving its collection; most health care workers are collecting these data and find this beneficial. The DHIS does not require collection of municipal ward-based health data, yet information is relevant only if it is used to solve local problems. Moreover, regular data analysis is essential to ensure that quality data inform decisions. The Amajuba Health District has undertaken a process through which contextual health data are gathered, shared, analysed and used for decision-making. Information is transformed into knowledge for action, but information must be current to be useful to managers at all levels.

The disaggregation of data contributes to the understanding of the population’s health, planning for their health needs and provision of targeted interventions in order to improve health outcomes, thus managing the scarcity of resources and providing financial risk protection for families and the NHI system. There is a need for the Department of Health to explore the possibility of disaggregating health data to a municipal ward and make this an integral part of the DHIS as aggregated data inhibit the understanding of the context and responsiveness to community needs, and, particularly in view of NHI, the timing is imperative. This innovation can contribute to increased responsiveness to community needs, thereby improving health outcomes and contributing to population health.

Disclosure statement

No potential conflict of interest was reported by the authors.