Hospitals Need More than Political Reform – Healthcare Quality v Spending pt 2

by Bradley Miller on November 19, 2009

My last post dove in to healthcare quality versus spending and introduced the Dartmouth Atlas as one of the key insight tools we have for healthcare spending and quality here in the US.  Essentially the Atlas takes a look at the entire country’s healthcare spending by using Medicare claim data as a proxy.  From this Medicare data the Dartmouth group can also glean quality of care data to pair with the spending figures.  And, for those of you wondering, yes, despite the fact that most Medicare patients are over 65, the data that can be gleaned from their data can highlight trends across the entire healthcare ecosystem.  Or, at least, it’s the best tool we’ve got.

I wanted to dive a bit deeper to look at regional differences in quality versus spending at a hospital by hospital basis. To take a look at the data provided by the Atlas and see if there were any high level trends that could inform the latest round of healthcare reforms.  I was wondering whether hospitals could provide high quality care in a cost effective manner.  In essence, the higher the quality of care, the better the patient outcome (and subsequently this saves the system money in the future from complications and further illness) – better outcomes for less money, aka a more efficient system.  Turns out it is possible to have high quality, yet more-affordable care.

Before I do that I want to recognize the limitations of the analysis of this work.  It hasn’t been blessed by a statistician or healthcare economist, etc.  It’s just looking at high level trends to show that, yes, there is pretty good evidence that quality healthcare doesn’t have to cost as much as it does today and that reform that properly aligns incentives can bring about not only savings but improved health outcomes.  In other fields, less spending often is equated with decreased quality.  In healthcare, this data indicates that it is possible to not only save money, but provide a higher standard of care at the same time.

Let’s break my analysis down by data type first.  I looked at overall spending per patient at hospitals during the last 2 years of that person’s life.  I’m not sure why the Atlas breaks the info down this way, but I was assured that this method does correlate to real world spending and is the easiest way to be consistent in collecting data.  That spending is reflective of the hospital’s overall spending habits.  The composite quality score measures how many patients out of 100 receive the right/highest quality of care.  So, if a hospital’s score is 85.4 that means that out of 100 patients that are seen in the hospital for a condition, 85 of them will receive the standard of care.

US Hospitals - total spending per patient in the last two years of life on the x-axis, and on the y-axis is an overall composite quality score.  Upper left quadrant is the ideal - lower spending, yet high quality care.

US Hospitals - total spending per patient in the last two years of life on the x-axis, and on the y-axis is an overall composite quality score. Upper left quadrant is the ideal - lower spending, yet high quality care.

For the country as a whole, you can see the plot of data for hospital spending (x-axis) vs quality (y-axis) – it’s quite a jumble, but all hospitals in the green quadrant are in the top half of being low cost (they spent less per patient), but are also in the top half of hospitals in terms of quality.  In otherwords, the hospitals in the green quadrant are low-spending, but high quality.  The opposite is true of the red quadrant – they are high spending, low quality hospitals.  It’s striking to see how many hospitals stray away from the top left part of the graph and it’s also striking how many high-cost, low quality hospitals exist across the country.  You can imagine that if all the hospitals in the red and white quadrants were able to change behaviors and become more effective and efficient just how much money the whole system could save.  And that’s not to mention massively improved outcomes.

To go a bit deeper with more meaningful numbers, I took the top 10% of US hospitals with low-spending/high-quality profiles and compared them to the remaining 90% of hospitals.  Overall, that top 10% spent $40,119 per patient, compared to $56,641 for high-spending hospitals for a savings of 28.9%.  The top 10% had a quality score average of 90.8, while the bottom 90% had a quality score of 85.7 – an improvement of 5.9%.  In summary, this means that the hospitals that perform at the top of the heap have a higher quality and they save, on average 28.9% of costs per patient.  Better outcomes, lower cost – it can happen, it does happen.

SF Bay Area Hospitals spending v quality graph.  Click to expand.

SF Bay Area Hospitals spending v quality graph. Click to expand.

However, it’s difficult to look at these stats on a nationwide basis because healhcare costs are more expensive (or less expensive) depending on where you live – just like every other living expense, healthcare is more expensive in expensive cities.  I took San Francisco (actually, the Bay Area) and Pittsburgh and subjected them to the same chart as I did the US.  The visual patterns are a little harder to discern, although in both cities you can still see that there are a handful of hospitals that have low-cost and high quality, while a majority are either high spending or low quality.  To me this means that there’s a lot of room for improvement across the board, even when looking at expensive and middle of the road metro areas.  While no clear pattern emerges, it is quite clear that most hospitals are not working as efficiently as they could be.  For you numbers oriented people, the top 10% in San Francisco saved 9.6% of costs and had 6.2% higher quality.  While in Pittsburgh, the top 10% saved 14.5% of costs and had 7.5% higher quality.

Pittsburgh Hospitals - spending v quality.  Despite being a lower cost city than San Francisco, overall both cities have similar graphical plots.

Pittsburgh Hospitals - spending v quality. Despite being a lower cost city than San Francisco, overall both cities have similar graphical plots.

There clearly are ways to save on healthcare expenditures, while still increasing quality.  Now, why is this in depth look appropriate for an entrepreneur blog that tends to focus on the biomedical sciences side of medicine rather than pure healthcare and policy?  Well, one reason is to help identify opportunities in healthcare for entrepreneurs.  Wherever there are inefficiencies in the system there is room for entrepreneurs.  However, the tricky thing is that it’s not so easy to just go start a hospital to introduce these changes de novo.  This change is going to require not only Congress, but the leads of hospitals to recognize the opportunities and to be entrepreneurial within their own hospitals.  Perhaps there’s even a market for efficiency tools from one hospital to the next – a sale of techniques that in the end save massive amounts of money and improve lives.  I wanted to point out these inefficiencies because I think they point to an opportunity for the right types of entrepreneurs to come in and make a difference.  I’m not sure how these ineffiiencies will be taken care of, but someone will figure it out.  I hope.  Put another way, if we leave reform only to politicians and insurance companies, more than likely we’ll have more of the same.

As a note, the higher the composite quality scores indicate not only better outcomes at that particular patient visit, but also tends to indicate a lower level of complications and ‘follow-on’ illnesses/diseases, thus saving the system a lot of money well in to the future in addition to the present.  Often these savings go unnoticed when thinking about healthcare reform.  The right care now not only amounts to savings in the present, but automatic savings well in to the future.

Leave a Comment

Previous post:

Next post: