Legislators around the world understand that the laws they make sometimes have unintended consequences. That is fine, but it becomes problematic when an unintended consequence erodes the very objective of the law.

This is sometimes observed with the Promotion of Access to Information Act (Paia) when access is restricted rather than promoted. As the National Health Insurance Bill, which is being finalised, considers information management, there is an opportunity to develop a new paradigm that would promote data sharing in a manner beneficial to all stakeholders.

Accessing data from the National Department of Health (DoH) can be tiresome, as many researchers, health journalists, civil society groups and ordinary people can attest.

The frustration starts when you try to explain exactly what you need at the enquiry desk. Routine data enquiries are normally dealt with efficiently, but if yours is more complicated you might find yourself down a rabbit hole.

It is not unusual to find yourself referred to a person who refers you to yet another person who in turn refers you back to the original person who then may ask you to submit your request by email. The resulting frustration is entirely preventable through a user-friendly and transparent data-sharing regime.

You could use Paia to obtain the information, but this comes with its own problems. First, it creates an adversarial engagement since it is often viewed as an accountability mechanism even when that is not the intention. Second, you may require the data in fewer than the 30 days granted by the act. Last, the department’s Paia manual may not be readily available, as required by the act.

SA Health department falls foul of access to information rules

These factors can form a huge disincentive for using Paia. Ironically, the act is sometimes used to prevent access to information rather than promote it, so it may not be the ideal instrument to access data.

Effortless and seamless data sharing is possible, as demonstrated during the Covid-19 pandemic, during which the department created a data-sharing infrastructure which made it possible for stakeholders to access data timeously, on various platforms and in user-friendly formats.

It meant the department was able to present data in real-time, in granular formats and packaged for use in different scenarios. Visual presentations were often used in communication targeting the broader public.

Raw data were made available to experts for use in further analysis and manipulation. This was so successful that organisations and institutions around the world used South African data as a reference point.

Moreover, regular maintenance of data through daily updates allowed stakeholders to track changes and trends and compare similar datasets, thereby enhancing the country’s credibility and standing.

The success of data sharing during Covid-19 can be attributed to two factors:

• The accurate scoping of the data based on the needs of data consumers. For this the department had to know exactly what data potential consumers may need and in which formats, taking into account considerations such as privacy; and
• The availability of the data. In contrast to a Paia data-sharing environment where data is made available on request, Covid data were available on demand. This contributed to data being available in real-time, making them far more useful and relevant.

While data sharing and communications regarding Covid can generally be viewed as a success, it wasn’t all smooth sailing. There was quite a bit of misinformation which had a negative impact, the effects of which are still felt.

Mind the info gap

While you cannot always completely counter the impact of misinformation, it can be minimised by being proactive and by staying ahead of the information wars.

With regards to the National Health Insurance (NHI) there are quite a few misconceptions floating around which can be directly linked to an information gap. This could be for a number of reasons including an inability to stay ahead in the information game.

One way to deal with mistrust and misconceptions is by implementing a comprehensive data-sharing plan for the NHI, if there isn’t one already. Such a plan should not be an afterthought, but an essential part of early planning. Not only would this be prudent from a planning point of view, it would also facilitate the early engagement of potential data consumers.

Visit Daily Maverick’s home page for more news, analysis and investigations

Section 34 of the NHI Bill, which deals with the national health information system, points to an awareness of the data-sharing obligation. Subsection 2 says data must be accessible to all stakeholders “legally entitled to such information”. A data-sharing plan as part of the regulations in the act will give substance to this provision.

While a comprehensive data-sharing plan will undoubtedly have to cover a wide range of topics, three areas of concern to data consumers deserve attention.

The first is access, and most consumers will agree it is the most difficult hurdle in their quest to obtain data. It starts with establishing the correct point of entry – it’s often not clear who to contact and how to make the request.

Then there are often challenges relating to authorisation or permission, which might involve waiting weeks or even months. Adding insult to injury, you might need to check intermittently on the progress of your request.

A final barrier to access comes in the form of a fee payable on request for the release of data.

Many can also attest to feeling utterly annoyed when the requested data arrive in an unusable format, such as a protected PDF document. Sometimes the format is readable only by special software. The worst-case scenario is data presented as a hard copy. 

‘Data is the new gold’, but without care is also a commodity easily tarnished

Presented with a non-user-friendly format, users often incur extra costs in terms of software and time spent on data extraction and transformation. 

Another concern is that the data request might be refused for reasons not immediately clear to the requester. This could trigger a dispute resolution process as laid out in the department’s Paia manual, provided it exists. This is not an ideal scenario since invoking Paia often leads to an adversarial relationship.

An on-demand solution

To address the concerns around access, the data-sharing plan should provide mechanisms for data to be available upon demand. With Paia, a request is made each time data are required, which places a heavy administrative burden on the department and can result in delays.

In an on-demand environment, data are packaged and available at the time of the request. Also, permissions and authorisation can be issued in advance to allow the consumer access at any time. In principle, access to data should be free of charge, which is achievable if the cost savings related to an on-demand system are realised.

To promote interoperability and the free exchange of data, the data-sharing plan should use formats based on open standards which must be clearly defined and published. No efforts should be spared to ensure data are available in various formats to avoid the need for specialised tools. Having only one format must be avoided at all costs.

The data-sharing plan should list all the available datasets and the conditions for access, thereby ensuring producers and consumers are on the same page and facilitating transparency. This will also foster trust between the parties and contribute hugely to a healthy data-sharing environment.

Digitisation has been identified as a major driver for the NHI. If achieved we can expect that it will produce large quantities of data which can be used to monitor, track and analyse performance with the view to continuous improvement.

Why South Africa needs a national Data Transfer Agreement for health research 

Stakeholders within and outside the government can contribute to this effort, provided that data are shared in a comprehensive and transparent manner.

Paia is not fit for this purpose since it engenders an adversarial relationship and is not responsive to modern data-sharing needs. DM

Dr Claire Botha is a health economist and policy analyst with a focus on public sector policy analysis, health financing and economics, costing and resource allocation. Her work spans more than 15 years, in diverse sectors such as government, non-governmental, higher education and research management.