Strengthening nutrition routine data using institutionalized health management information systems for decision making: analysis of best practices and lessons learned from implementation in Burkina Faso | BMC Health Services Research

I. Participants’ characteristics

In 2022, a total of 64 key informants spread over 2 health districts, 2 regional hospital centers and 2 health regions, who were involved in the enhancement of nutrition information, participated in individual interviews. Concerning the different roles of respondents, key informants, the health facility workers are involved in generating data from their frontline service provider roles. The data managers at district ensured the entry of data in DHIS2 and the quality assurance of data. At regional, the actors ensured the quality control to ensure the avaibility of quality data (Table 1).

II. Results of the capitalization study conducted in 2022 on best practices, lessons learned and conditions for sustainability and scaling up

The results of this study showed that the National Nutrition Information System (NNIS) Improvement Project has strengthened the availability of nutrition data in the DHIS2 and the analysis, quality, and use of routine nutrition data to improve the performance of maternal and child nutrition programs. The main themes highlighted by the respondents based on promising practices are presented in Table 2.

II.1 Observed changes, success factors and challenges related to data availability

Introduction of new nutrition indicators in the DHIS2

The results of this study show that the project has enabled the effective integration of nutrition indicators (prevention and care) in the DHIS2 by filling the data gap. In 2018, following a review of indicators, a consensus was reached on 13 new key nutrition indicators that were retained in the small-scale phase. The challenge was to not select “too many” indicators, which could hamper the data quality and workload of service providers. Only tracer indicators that allowed proper monitoring of the implementation of the national nutrition plan were selected for the test phase.

At the end of the small-scale phase, two indicators, namely the number of health workers trained on the IMAM protocol and the number of community-based health workers trained, have not been validated for integration into the national collection tools, as they can be extracted separately from activity reports.

Among the success factors for the introduction of the new nutrition indicators, we noted (i) the leadership and high-level commitment of the Minister of Health, (ii) capacity building of health workers, (iii) a participatory approach throughout the process with the involvement of health districts and key partners, and (iv) search for consensus with exchanges between the implementing actors.

However, the 2019 strike by health workers disrupted the implementation of the small-scale phase.

Additionally, during the process of selecting indicators as part of the overall review of the primary tools of the DHIS2, there was a preference for curative interventions over information education communication.

Data from the vitamin A supplementation (VAS) campaigns in the DHIS2 were also set up and integrated into the DHIS2 instead of Excel files. Some respondents pointed out that this contributed to improving the archiving of VAS data through DHIS2 and the traceability, availability, and quality of VAS data. Dashboard extractions show shortcomings in the quality of the VAS data, with coverage beyond 100% in some districts.

II.2 Observed changes, success factors and challenges related to routine data quality

The results of the study showed an improvement in data quality in term of completeness and promptness rates through the following key actions: (i) decentralization of data entry in DHIS2, (ii) the organization of on-site data quality control as well as validation of data at the district level and (iii) the organization of nutrition performance reviews.

Regarding decentralized data entry, the use of tablets versus paper has greatly improved the quality of data in terms of timeliness, completeness, and internal consistency. This graphic 3 shows the comparative analysis of the completeness and timeliness of monthly reports in an intervention (Yako) and control (Ziniare) health district (Fig. 3).

The existence of the electronic consultation register (REC) with tablets and the training of health workers were factors that facilitated the implementation of decentralized data entry. However, the instability of the internet connection and renewal of data bundles are major challenges for their sustainability.

The data quality control at health facility level with the new process introduced by the project, made it possible to compare the nutrition data from the monthly reports with the data from the registers and to correct the aberrant values. The same is true for the holding of data validation sessions at district level for ensure data quality.

II.3 Observed changes, success factors and challenges related to the systematic analysis of data

Respondents highlighted an improvement in routine data analysis with the development of a dashboard in the form of graphs, tables, and maps; the organization of performance reviews with a systematic analysis of key indicators for monitoring the performance and identification of bottlenecks; and the development of a problem-solving plan. This performance reviews organized at regional level in attendance of data managers, nutritionists, pharmacists and health workers. One of the interviewees declared “With the dashboards, we see a greater appropriation of nutrition issues. The actors are more attentive to the evolution of the indicators”.

These dashboards, present at the national, regional, and district level, have further decentralized to the level of individual healthcare facilities, significantly reinforcing the feedback mechanism (Fig. 4). This Fig. 4 below shows an overview of dashboards of key nutrition indicators in DHIS2.

Among the success factors that facilitated the systematic analysis are (i) capacity building of actors, (ii) existence of a performance review guide with a directory of key indicators, (iii) availability of qualified staff at the national level for the design of dashboards, and (iv) active participation of key actors. The main constraints encountered were the lack of data validation workshops at the health district level, COVID-19, and the security context.

II.4 Observed changes, success factors and challenges in using data to improve program performance

Factors contributing to the successful use of routine data to inform the national nutrition plan include: (i) the development of a public interface that expands access to nutritional dashboards, (ii) the conducting of nutrition performance reviews across all 13 regions and at the national level, which have contributed to improving the quality, analysis and use of data; (iii) dissemination within district health councils, regional nutrition councils, and during Vitamin A Supplementation review meetings.

Conversely, barriers hindering the use of routine data for decision-making encompass insufficient policy briefs, limited financial resources, and insufficient recommendation follow-ups.

II.5 Lessons learned from the implementation that are relevant to other studies and projects

The experiences from Burkina Faso have yielded valuable insights such as:

• Conducting a small-scale phase to test the indicators is an important step to take prior to scaling up.

• A participatory approach, with the involvement of all actors at different levels is an important point for the introduction of nutrition data in the DHIS2. It enabled a consensus to be reached with stakeholders on the key indicators to be included.

• The absence of standard definitions for routine nutrition indicators does not facilitate an informed choice on the minimum list of indicators to be monitored routinely.

• Specific data disaggregation by gender were overlooked, which may limit the ability of decision-makers to use the data to address gender equality issues.

• Advocacy is important to integrate prevention indicators into health facilities in a more curative-oriented health system. The tracking preventive indicators by the health workers allowed to put emphasis on the prevention. In public health, the prevention came first than the treatment.

• Beyond the inclusion of indicators in DHIS2, special emphasis should be placed on data quality according to the recommendation commonly made by respondents.

• The decentralized entry of data is a best practice that improves data quality in terms of timeliness, completeness, and internal consistency.

• Implementing a weekly IMAM data collection protocol has effectively bolstered nutritional surveillance during crises.

II. 6 Conditions for sustainability, reproducibility and strengthening of the national scale-up

According to the interviewees, the initiative was sustainable as the Ministry of Health carries out the activities, the project is integrated into the HMIS. In addition, parallel data collection with excel files has been gradually abandoned in favor of DHIS2. However, to strengthen the projects’ sustainability, to strengthen the system, the maintaining information systems requires continued investment, a certain number of actions are necessary:

• Continuously enhance the expertise in the Nutrition Information System and integrate training needs into training curricula in health schools.

• Organize supportive supervision to update actors in the field.

• Increase financial resources to bolster logistical capacities in electronic equipment (Smartphone or tablets).

• Ensure the application is regularly updated.

• Strengthen the dissemination of dashboards in existing frameworks.

• Strengthen the regular holding of data validation sessions at the district level.

• Have a stable telephone network and internet connection.

• Regarding the weekly IMAM collection, data is currently collected via telephone and Excel files. Transitioning to tablet-based data collection could enhance the archiving process.