Two things that have been very clear from the last 18 months of campaigning against the government's changes to the NHS: there are a lot of people opposed to the changes, and there are a lot of different groups campaigning. It is reassuring that so many people are against the government's policy. It is also powerful that there are so many groups, because each group has a different skill and can fight the changes in their own way. The problem comes when several groups decide to do the same thing: they duplicate effort (wasting their own scarce resources) and it can mean that the impact of their efforts is diluted. As we move into the next phase of opposition to the Act, we need to make sure that many different groups can be involved, independent of each other, but through mutual respect and sharing, they can be stronger than as a single organisation.
From now until the next election (at the latest, May 2015) everything that is wrong with NHS treatment in any part of England will be blamed on the Health and Social Care Act. In some cases this may be unfair because it may be an issue not effected by the Act, but the government has taken an attitude that everything that is good in the NHS is due to their policy, so the opposition to the H&SC Act are merely playing by the same rules: this is politics. Various organisations, big and small, are now looking into collecting information about the NHS with a view to using this information to attack the policy. So groups intend to collect information about contracts with the private sector under AQP (Any Qualified Provider), out-and-out privatisation of services, cuts in services and cases when patients received a poor, or even dangerous, service. Each of these different groups have a different focus, and each one has a "membership" with the skills to provide the information needed. On its own, each of these projects can have an impact. However, together they become really powerful.
When one group makes an effort to collect information they are understandably reluctant to allow others, who have not put in the effort, to use the information. I understand this reticence. However, when done correctly data sharing is powerful because other groups will look at the data from a different angle: some may look at the financial aspect on the taxpayer, others may scrutinise the corporate governance, others may look from a clinical perspective and others may look from the patient's point of view. All are valid, all are useful, and all need access to data.
Sharing of data does not have to involve a central organisation. Indeed, it is important that there is not a central organisation, not least because that represents a central point of failure: if the central server goes off line, no data is available. Instead, we need there to be many databases of information. However, the significant point is to ensure that data is shareable. That is, there is a machine readable description of the data, and the dataset can be read, in part or as a whole. Whenever a group decides to build an online database one of the key questions they should ask is "how can we share the data?". The single central organisation that is required is a list of the data available. Such a website would be simple: for each entry there would have to be a link to the description of the data, a link to an acknowledgements page (the user of the data must indicate whose data they have used) and a link to access the data. The data could be a web service, a raw csv (comma separated value) or XML file, or even through a Google documents spreadsheet. The data access could be to all of the data, or there could be features to allow it to be filtered. But what is most important is that the data access must be open and the data must be documented.
The power of such an approach is that it enables mashups. A mashup is where a web developer takes data from different sources and combines them. So information from a geolocation source (for example openlylocal.com) can provide constituency or electoral ward data; a database of local election votes (for example, from a local authority website) can give the majorities of the local politicians; and Hospital Episode Statistics can show the use of an NHS service in a region. All of this data could be mashedup with custom data showing, say, the volume of complaints collated from various sources (local press, NHS Choices, voluntary organisations) since a service has been privatised. The result would be a warning for local politicians about how vulnerable they are to candidates who stand on a platform to restore the NHS. The organisation who collected the complaints would still be able to use their data to make the point they want to their audience: that complaints increased after privatisation, but the mashup with other data gives another message, to a different audience.
The open source aspect of such data does not weaken the message of the organisation collecting the data. Instead it realises the potential within the data, and makes it more powerful. This is a stone soup approach: everyone contributes what they can and we all benefit from the result. In our case, each organisation collecting information make it available so that others can use the data in their own way, and ultimately, we will all benefit from our efforts by restoring our NHS to a collaborative and public system.