Yes, a majority of us do skim through the headlines and move on. It's a guilty habit that leaves us with incomplete or even inaccurate inferences- a habit that has been called out especially in the last few years. We can see a significant enough population who are actively trying to keep themselves informed, but failing- it takes time and dedication. Even if news articles are read, a study from NNGroup found that 79% of articles are scanned. From a UX point of view, one could argue that there's a failing in user engagement and we in the design industry could work to encourage that initial curiosity.
There are 2 proponents that one could argue are in the way of this engagement.
Firstly, headlines (or the design articles links attached to the CTA) need to prompt users to click and engage and news websites achieve this, because they need users to click as many pages as possible. But websites that have social aspects to them like Facebook and Reddit aren't focused on that. For these websites, the user needs to summarize, and the link to the article is treated as little more than a footnote. These websites are designed for users to stay on the site for discourse, and it is not within their interests for their users to leave their sites. For the source news websites, I can suggest promoting thumbnails with more information, rather than just a headline and a small tagline that prompts further reading. Further, the goal is to get people at least on the article site, and my suggestion as a UX designer for news sites would be to promote a synergetic relationship with these sites.
Secondly, as NNGroup noted in the same study mentioned above, text walls written objectively, concisely, market-ably and formatted for scanning yield a significant rise in engagement. Blinkist, a service that aims to help you summarize whole books in 15 mins does this very well. They write and divide their content in digestible chunks with bulletted content, a read-along narrator, large and legible text and with the aim of making it easy and enjoyable to move on from one page to the next. Blinkist makes content enjoyable to read- and breaks up content to be read as a full narrative. We can aim to achieve this with articles- we can aim to make articles a joy to read. Obviously the context is different, but the intent, the initial curiosity is the same, as well as the end result.
Obviously there are bigger parts moving in the background of this problem- but at the capacity that we can solve it, we should advocate for a better ability to be as engaged with the world events as we obviously want to be.
Post-script Note:
I've since decided to try and create a mobile news aggregator app that seeks to solve reader engagement here.