Wednesday, December 2, 2015

2015 In Review

It’s now December and as each year ends, I always look back on the challenges and achievements of the past 12 months.    Here’s my sense of 2015.

ICD10 - billions were spent, countless other projects were delayed, and the transition occurred on October 1 without a major incident.   We’re monitoring daily cash at all our hospitals and there has not been significant impact on denials, payments, or discharged but not final billed accounts.    Did we get our money’s worth?    I have argued and will continue to assert that ICD-10 benefited no one.  The diagnoses used are more variable so there is less precision in their use.  Clinical documentation (in general in the industry) does not have the specificity needed to justify the more granular ICD-10 codes.    The notion that quality measures can now be computed more accurately from ICD-10 coded administrative data is just not true.     The right path is to plan for a future in which fee for service is replaced by bundled payments so that ICD vocabularies do not need to be used at all for billing.    Natural language processing  will be able to turn unstructured text into SNOMED-CT coded observations to support analytics.    I know that ICD-9 is obsolete and did not include many modern concepts.   However, we should have saved our billions and waited until natural language processing and SNOMED-CT was ready (or a convergence of SNOMED-CT and ICD ideas such as will be implemented in ICD-11)   The end result of years of work 2012-2015 is that many IT stakeholders think IT was distracted by projects that added little value.  The good news is that now that ICD-10 has passed, we can return control of IT priority setting to customers.

Meaningful Use - Stage 2 was revised and Stage 3 was finalized with a comment period.    We can only hope that the comment period convinces CMS and Congress to shift the Meaningful Use program into a merit-based payment incentive program, acknowledging the Meaningful Use has achieved its goals.   Just as with ICD10, we need to turn the IT agenda back to customers - patients and providers - who want improved quality, safety and efficiency.   As we’ve seen with Stage 2, it is too early to propose a Stage 3, because we do not really know what has worked in Stage 2.   I have advocated for moving to an outcomes approach.    If you want to give Apple Watches to all your 80 year olds to monitor their exercise patterns and support a patient-based medication administration application on the watch, go for it.   If you want to hire high school students with clip boards to visit elderly patients and do home checks, go for it.   The outcome might be better health and fewer hospitalizations.   The tactics should be up to patient centered medical homes and ACOs, not regulation writers.     My secret hope is that CMS decides to remove the penalty phase of Meaningful Use, enabling every EHR vendor to ask their customers - should we spend the next 3 years implementing the Certification rule (which is voluntary) or just ignore the entire Meaningful Use program and innovate to accommodate the needs of alternative payment models?  My guess is that the majority of hospitals and professionals would tell vendors to abandon the certification effort and focus on value added enhancements.   At that point, the Meaningful Use program could be considered a success and be moved into a historical status - still on the books, but not pursued by most.

HIPAA Omnibus Rule - In 2015 we were told to share more data with more people for more purposes, but to never allow a single byte to go astray - an impossible task.    OCR stepped up HIPAA Audits and enforcement at a time when threats from cyberterrorists, organized crime, and hackivists peaked.    Rather than focus on all the vulnerabilities and the mistakes made as documented on the wall of shame, I’m hoping that as a country we can focus on the positive - working together as a society to identify the real threats and collectively take action to mitigate risk through policy, education and technology.    We need to stop creating a climate of adversity among regulators, providers and IT departments.    Later this week, I’m meeting with National Security leaders from  the Department of Justice.   I can only hope they will propose a collaborative, positive approach.

Affordable Care Act - 2015 was a year for big data analytics.   BIDMC’s ACO was the #1 ACO in New England and #3 nationally, in part because of our ability to aggregate a common data set of clinical and financial data from 26 EHRs (all our loosely affiliated clinicians).   We used this data for care management, quality analytics, and benchmarking.   We saved Medicare $50 million in 2015 alone.    It’s clear that turning data into wisdom through the use of novel visualizations, alerts and reminders works.    Meaningful Use did not tell us to do this.   The Affordable Care Act told us to achieve an outcome and innovation happened because incentives were aligned to motivate us.

Cloud  -  2015 was the year in which the cloud became a viable option for just about every application in healthcare.   Amazon and Google both agreed to sign business associate agreements.    Many companies offering cloud-hosted services agreed to indemnification clauses for privacy breach.   At this point, the cloud can be more reliable, more secure, and more agile than local hosting.    Pilots for BIDMC include moving our development/test environments and disaster recovery to Amazon.    Production systems are likely to follow.

There you have it - 2015 has come and gone with major federal programs winding down and control being returned to the private sector.   I’m incredibly optimistic about 2016.   As I’ll write about soon, our agenda is filled with new ideas and it feels as if the weights around our ankles (ICD10, MU) are finally coming off.


Kevin Groff said...

I would like to know how your DR scenario goes with Amazon. I take it there are no mainframe or midrange computing platforms at BIDMC. If so and you separate those apps from your x86 from a DR perspective, will there be an integration impact in a real DR scenario? I do like the cloud model, in principle, for DR but do you believe a hospital needs to be on all x86 prior to investigating that option?

Georgia Newman said...

Thank you for your words of wisdom. I hope that your 2016 will also include more hours of sleep and less blogging at 3am. Aren't you wearing your fitbit?

I am a solo physician truly suffering with all of this. I finally got my MU dashboard acceptable, but hours and hours of time were spent changing LOIN-c codes, moving falls risk from one spot to another, etc, etc. Stupid, stupid, stupid.

Wrote a humorous blog for Physician's Practice titled "CMS, the Quagmire" about my, as still unsuccessful, attempts to register to view my 2014 PQRS stats. Although I am a proponent for a single-payer system, every contact with CMS has been a disaster; Maybe a single payer system OTHER than CMS.

Georgia Newman MD FACP
Oberlin Internal Medicine
Oberlin, Ohio, 44074