This post continues my review of the online course, Using Data in the Health and Social Care Environment. The second part of Section 2 is about getting data from websites.
First there’s a list of websites, some with screenshots and some without, some with summaries and some without. The focus is on Public Health England, but several other websites are noted briefly.
Then the remainder of this material relies on short videos on YouTube instead of text, and it consists entirely of two case studies.
These case studies, like the previous case studies in the course, are both about getting more public funding for the third sector.
Case study 1
The first case study is about getting more public funding for the third sector so as to encourage healthy lifestyles in Burnley.
Setting the scene with secondary evidence
The video, Setting the scene with secondary evidence, is a brief demonstration of how to find some data in New Economy’s Unit Cost Database, which is a spreadsheet, not a database. And then how to find a page on the NHS Choices website. This didn’t really need to be a video.
New Economy is part of the Manchester Growth Company, which seems to be an association of sixteen other companies, all but one sharing the same address and phone number.
The video starts showing a Firefox web browser displaying the New Economy website without telling you how to get there. Then it makes clicking on a link sound so difficult that you need a video demonstration. It assumes you have spreadsheet software that can open the Microsoft Excel file. I opened it in LibreOffice, which worked just fine.
Once the demonstration is in Excel it moves very fast. For example, the voiceover says “And find the Health tab” but the video moves on quickly before you can really see where the Health tab is. Anyone new to spreadsheets will take longer than this to find the tabs at the bottom of the window.
Then it says, “Expand the Hospital inpatients average cost per episode elective and non-elective admissions parent group”, without telling you anything about how to do that.
Finally you read a figure from the mysterious “final column”, which the video claims was £3,887 at the time of recording (even though it’s now £3,987). The spreadsheet file says it was last modified in March 2015, and the video was uploaded in June 2015, so it’s just possible the figure changed at around that time, although further investigation suggests otherwise.
Hidden off-screen, there’s a column that claims to be a link to the original source of the data, but the link is broken. On tracking down the report by NICE, it turns out it was updated in June 2015 two weeks after the video was uploaded. The cost there is £18,300 over five years, which averages out at only £3,660 per annum. (The costs might not be the same in each year following a stroke.) It looks as if New Economy has adjusted the numbers in a way that’s hard to explain.
But the NICE report also reveals that it’s not the original source either. The source it cites is the 2008 version of the clinical guideline for stroke, which was last updated in September 2015. I can’t find any costs in the guideline, but it says there’s now a tool “to estimate the local costs and savings involved”. I tried to find the tool, but that link is also broken.
So it looks like the figure of £3,660 came from NICE in 2008, which means it was most likely based on 2007-08 costs. But in the video the date of the costing is wrongly set to 2010-11 (you can see it in the orange box to the right of £3,660). If you set the date correctly to 2007-08 that box turns red, probably indicating “Wow! That was a long time ago. This cost is probably unreliable by now!” And the cost we’re looking for changes to £4,309 this year (2015-16) after adjustment for inflation since 2007-08.
All this assumes that stroke care hasn’t improved since 2007, so that the only thing affecting costs is inflation. In reality, the field of stroke care is vibrant with research and development, so the assumption is almost certainly wrong.
It looks like whoever made the video didn’t set the date correctly, misread £3,987 (which was wrong anyway) as £3,887, and didn’t check New Economy’s assumptions.
It also looks like the proper way to estimate local costs for stroke care is to use the latest official tool (if you can find it), not the slightly mysterious figures from New Economy.
For the NHS Choices part, the video does tell you how to find the website but not how to find the page on stroke prevention. It’s pretty easy to search for it, though.
Building the case with data
The video, Building the case with data, demonstrates using NHS England’s Commissioning for Value Explorer tool and finding a correlation. The tool uses Flash, which my computer says is a security risk.
The voiceover confidently says, “When you first access the tool, it will look like this…” Mine didn’t look anything like that. Even the buttons are different. The differences didn’t prevent me from following the video, though. Subsequent steps worked.
I later discovered that there are (at least) two versions of the tool. The course confuses them, sometimes referring to one and sometimes to the other. This confusion affects both the course content and the quiz questions. The other version of the tool is here: Commissioning for Value Explorer Tool
Eventually you find there was a positive correlation (0.77) between mortality from stroke and an index of socioeconomic deprivation back in around 2010. The voiceover points out that this correlation doesn’t mean that one of the factors causes the other. But then immediately goes on to say that the correlation is strong data to make the case (for the CCG to provide more third sector funding), without explaining the connection.
In reality some third factor might be influencing both mortality from stroke and socioeconomic deprivation, and much might have changed since 2010, but the voiceover doesn’t consider these possibilities.
Also, the target of the proposal, East Lancashire CCG, is pretty much in the middle of the scatter plot. Its deprivation index and its stroke mortality were both middling in 2010. This makes it look as if stroke mortality and the effects of deprivation might not be high priorities for the CCG unless things have worsened in the last five years. The voiceover doesn’t consider this either.
Next, the video moves on to Public Health England’s Health Profiles tool. Again, the video is out of date. It uses the profile for 2014 but the current profile is for 2015.
After summarizing some of the information in the profile (Burnley is worse then the England average on lots of measures) the video ends abruptly.
Using strategic documents to link the case to local priorities
The video, Using strategic documents to link the case to local priorities, uses the The King’s Fund’s Health and Wellbeing Board Directory to locate a document summarizing the Health & Wellbeing Strategy for Lancashire. But the King’s Fund links to an old document from 2012, which talks about “the emerging Health & Wellbeing Strategy”. The actual strategy direct from Lancashire County Council is dated 2014. The text is much the same, though.
I think it’s probably better to use normal Internet search to locate documents, instead of the King’s Fund’s outdated links.
The video moves on to the East Lancashire CCG website, and again it has changed since the video was made.
Finally the video goes back to the King’s Fund map for a link to Lancashire’s Joint Strategic Needs Assessment (JSNA). On the JSNA website the video refers to “item six”, but there are no numbered items shown in the video or on the current website. The video then displays a quote without attribution. Google couldn’t find the quote in any document anywhere.
The video then goes to another page on the JSNA website, and shows another quote, without attribution, that’s not on that web page. I suppose it might have been when the video was made.
The concluding video in this case study summarizes what’s gone before in two sentences, which takes 30 seconds and didn’t require a video.
Case study 2
The second case study is about getting more public funding for the third sector so as to improve health outcomes generally in Suffolk.
There’s only one video, and it has similar problems to the preceding ones, which I’ll not document in detail. For example, the Suffolk data is from 2009.
It looks at predicted future health problems, comparing rural and urban areas, and concludes that the urban areas were going to have more problems. Six years on, did predicting those problems in 2009 mean that things changed in Suffolk? I don’t think we’re meant to ask that kind of question.
I spent so long watching these videos (pausing them often to do other things) that my session timed out…again.
Next, there’s a quiz on the content of the videos. The introduction says there are five questions, but there are really ten, and each question has several parts. Each question asks you to find some information on a website. This soon became tedious, and I gave it up after a while.
When I returned to complete the quiz, it had lost all my previous work and I had to start again.
Some of the problems with the videos affect the quiz questions too. However, most of the information you’re asked to find is very old, and therefore unlikely to change.
All the quiz questions can be answered without any knowledge of the course content. You simply follow the instructions in the quiz itself to find the answers. This demonstrates how little use this part of the course is.
The case for change
Watching these videos did make me think about how to make a case for change in health and social care. The main lesson from the videos is that third sector organizations often change nothing. Their bids for funding tend to make a case for the status quo.
The sequence of events is roughly this: research into some problem or other was funded in around 2008, the results were published in around 2010, and commissioners responded with a strategy in around 2012 which basically says, “something must be done.” Then anyone can come along and ask for money for any activity loosely related to whatever problem was identified back in 2008.
Bidding for funding isn’t about making a case for change. It’s pushing on the open door of current commissioning strategy. The decision to address that particular problem was made in 2008 and everything else follows.