Offshore Coding: Week 1: Chaos

EDITOR’S NOTE: Dr. Anon is a U.S. physician based overseas who wishes to remain anonymous.

Following implementation of ICD-10, what has happened in the coding world as it pertains to outsourcing? The answer is simple: chaos.

For those of you old enough to remember, I am not describing a Get Smart episode with the international organization for evil known as KAOS – but the scenarios may mirror each other. There are now many coding vendors in India, the Philippines, Jamaica, and South America – put simply, although some American providers may have been listening to industry predictions of 40 to 50-percent drops in productivity and thus hired staff to train and deploy, others turned a deaf ear and did not heed the warning. So, what is currently happening?

1  Staffs in India (where labor laws are very loose) are being forced to work overtime and at least one weekend day per week. This may not be happening in the other outsourced countries, as labor laws may be tighter.

Impact analysis: Extra volume is everywhere, but due to mental and physical exhaustion among coders, a drop in quality inevitably will occur. How long can this mandated work schedule be sustained?  Many coders seeing a favorable market have left vendors that are forcing this type of work schedule, leaving inexperienced staff with less than six months of experience to handle the workload.

2.  Everyone, including auditors (a second set of eyes, and not really the definition that is widely known in the U.S.) has been deployed to meet client expectations.

Impact analysis: There obviously should be some oversight of coding quality, especially for those coming right out of training.

3.  Vendor ICD-10 implementation coincided with acquirement of temporary/new ICD-9 business for those clients trying to get their internal staffs acclimated to ICD-10.

Impact analysis: There has been very little practice time on dual coding by vendor staffs due to the onslaught of new ICD-9 business. Therefore, unfamiliarity with subject matter during ICD-10 implementation, and without available QA staff, can create a massive impact on coding quality.

4.  The outsourced staffs are utilizing outdated books (2014) and therefore erroneous codes or incorrect exclusion 1 exclusion 2 applications are being followed.

Impact analysis: Wrong codes or not enough codes are being applied to the claims, and this too is creating a drop in quality due to inadequate oversight.

5.  In order to meet some expected production, the coders are hurrying through the charts and not reading the documentation, which will have a deleterious impact on the back end.

Impact analysis: Clients should be prepared for massive denials. This will mean that the claims appeal department and the billers will be putting in many extra hours of work to get the claims paid appropriately.

6.  In order to meet volume, codes such as external cause codes are not being assigned on the claims.

Impact analysis: Although there is no national requirement for mandatory ICD-10-CM external cause code reporting, if there are state-imposed, payor-based requirements regarding the reporting, these will not be met and data capture will be incomplete.

7.  Chronic conditions and history codes are not being captured on the claims.

Impact analysis: Not capturing the patient’s total clinical picture will impact case mix with a drop in severity of illness. Even a small change in CMI for IP might send the financial department into “suicide watch” because of the large impact on hospital revenue.

8.  Whether by client design or coders rushing to get their quotas finished, many claims are being released with unspecified codes.

Impact analysis: This creates the possibility of denied claims for lack of specificity. Although the Centers for Medicare & Medicaid Services (CMS) has a moratorium of one year on specificity within a family of codes, this guidance only applies to Part B physician professional claims and not facility billing. Remember that the government dangled the carrot to get the providers on board for ICD-10 implementation.

9.  There are huge backlogs with no end in sight.

Impact analysis: Reimbursement will be held up. Look at your DNFB and start to seriously question the vendors, forcing them to give you an action plan, because these backlogs will cripple your cash flow.

10.  Claim submission filing limits are not being met.

Impact analysis: Denials abound. Make sure that you are working with the vendor to code claims with the highest dollar and shortest filing limits first. I don’t have to remind you that if you do not meet filing limits, you have no appeal rights.

There were some facilities that were proactive and forced the offshore vendors to prove that they had the resources to meet the ICD-10 expectations. They asked for detailed lists of coder names and they tracked attendance and short- and long-term leaves. 

Why? In short, it was to prevent sharing of logins and account information for individual coder productivity. Were they micromanaging? Absolutely, but it was necessary to protect the reimbursement expectations. Those groups that left it up to the vendor to have a plan for ICD-10 without following up were not doing due diligence, making sure that projections were meeting the expected deliverables.

As a facility or provider group that has outsourced coding, you need to stay on top of what is happening to get the results that you are expecting. Don’t take the vendor’s word for it that what is expected is happening, because transparency may not be in place. Ask for a quality score on every single one of the coders assigned to your specific project. Run reports for codes expected to be found on claims such as accidents. Run reports on unspecified codes. Get involved in the project weeds, and don’t just accept an overview that everything is smooth sailing, because it is often not! 

I hope that those of you that have outsourced your coding have also established a line of credit with a financial institution, because you may need it in the upcoming months.

You don’t want to be caught off guard if something drastically goes wrong.

Comment on this Article

Build an ICD-10 Rapid Response Team for Revenue Emergencies

Now that ICD-10 is a reality in U.S. healthcare, providers are finding that Oct. 1, 2015 didn’t bring with it the doomsday scenario they thought it might – but they also know that ICD-10-related challenges that aren’t apparent now may manifest themselves in the coming months. 

The latest data from the Workgroup for Electronic Data Interchange (WEDI) showed that larger providers felt confident they would be ready for ICD-10; readiness, however, is a subjective term. No doubt most providers were ready to submit claims, but anticipating the denials and payor preparedness yet to come constitutes another level of readiness. 

ICD-10’s enhanced coding guidelines could lead to a higher incidence of denials due to coding errors. What is more likely is that ICD-10’s greater granularity will result in lower reimbursements if documentation doesn’t provide adequate detail to justify higher-severity coding and DRGs. 

Providers that were truly ready for ICD-10 have conducted end-to-end testing. This means they have submitted claims both based on ICD-10 documentation guidelines and native coding in ICD-10 and have received adjudicated claims from payors. They have analyzed coding effectiveness and clinical documentation adequacy, and where they found gaps, they provided coders with additional training and improved documentation. They found their risks, analyzed them, and addressed them. 

Do you need an ICD-10 rapid response team? For providers that haven’t been able to get to that level of preparation, the risk of significantly decreased cash flow exists. Such providers need a rapid response team watching for coding and documentation gaps and a plan to address those gaps as they are identified. The following are four essential steps that organizations will need to follow to build an effective ICD-10 rapid response mechanism:

First, make sure health information management (HIM) and revenue cycle data are easily accessible. Most organizations tracked and reported their top 25 diagnoses and top DRGs under ICD-9, and they need to be prepared to track and report the same data for comparison under ICD-10. Other important data points to track and compare include accounts receivable (AR) days, cash on hand, discharged-not-final-billed (DNFB), discharged-not-final-coded (DNFC), and case mix index (CMI). Organizations should create a dashboard to track each of these key performance indicators. 

Second, develop a team of experts that can identify risks using data. While a typical command center approach often seen with large rollouts of enterprise systems may not be necessary, having focused teams aware of key indicators of risk is a prudent strategy. Billing experts, knowledgeable coding staff, and denials management experts would be an ideal mix for a rapid response team. Additionally, ensure clear lines of communication with payors to facilitate remediation of unforeseen impacts to revenue. 

Third, determine the thresholds at which the ICD-10 rapid response team takes action. Facilities need to determine whether a metric fluctuation is an anomaly or true cause for alarm. Certain areas, such as claims data, coding audits, and case mix, may need 14-60 days of data to determine whether to take action. Other areas, such as clinical documentation improvement (CDI) queries and denials, may need only a few days of steady differences to identify if a problem exists.

Fourth, determine how the team will respond. With a rapid response team in place, data at its fingertips, and applications at the ready to identify risks, organizations can act in an effective manner. In general, prepare a dedicated workflow pattern to follow once an issue is identified. Identify key accountable individuals with the authority and ensure their ability to manage change.

No matter how rapid, the response must be right. With decision-makers in place,organizations should provide resources to dedicated denial management teams, clinical documentation improvement teams, and coding improvement teams in order to get their revenue back on track.

Rapid responses to manage denials: In the early days of ICD-10, it may be useful for organizations to review their top ICD-9 denial reason codes and the associated CCs/MCCs, and monitor these daily for trends of increased denials. A mix of people, processes, and technology is always in order when it comes to limiting denials. Organizations should ensure that they have adequate staff, efficient processes, and capable technology in place to address increased denials in a timely manner. 

Rapid responses to improve clinical documentation: In a perfect world, all physicians would be up to speed on ICD-10, and CDI specialists already will have worked on high-volume, high-impact DRGs that are affected by ICD-10. If that is not the case, CDI activities can serve as validation to providers about the education they have received. They may fall into old documentation habits, but physician queries and discussions with physicians about documentation concurrent with a patient stay can educate physicians in real time about new documentation guidelines for certain DRGs. 

Rapid responses to improve coder productivity: Coder productivity, typically measured by the number of charts per day an average coder can complete, must be tracked. Most organizations know what their baseline productivity is; if not, it can be calculated easily with data from most coding systems. Some organizations that have prepared for ICD-10 with dual coding saw minimal productivity loss, while others saw productivity hits similar to what was predicted: 20 to 50 percent. The hit to revenue caused by a 25-percent loss in productivity could be significant.

Hiring more staff coders will be difficult at this point, as will be securing contract coders. With the ubiquity of remote coding, organizations are competing regionally, nationally, and even internationally for qualified, experienced coders. One solution is to boot-camp train entry-level coders. Give them the training they need to code the simpler outpatient encounters such as labs and radiology, then shift more experienced coders into more difficult inpatient coding scenarios.

In conclusion, most if not all provider organizations could benefit from maintaining an ICD-10 rapid response group. By continuously analyzing data and deploying resources to plug revenue gaps, providers can remediate ICD-10-related challenges.

At the same time, don’t forget the “blocking and tackling” activities essential to ensuring accurate, timely revenue. One such activity is a documentation and coding review. Organizations should audit 10 to 20 percent of their charts over six months. Why six months? That’s when organizations should start seeing improvements. Another basic activity all organizations should engage in is reimbursement analytics. Comparing revenue by DRG, year-over-year, beginning with October 2015 by comparing it to October 2014, is essential. Doing so will help organizations find major shifts in reimbursement.

About the Author

Warren Hansen is an associate director of provider consulting at Optum 360. Warren was one of the principal designers of Optum’s ICD-10 Consulting solution. His extensive and diverse healthcare background includes IT management, revenue operations, HIM, and project management.

Comment on this Article

ICD-10: Smooth Transition for Claims, PMS Vendors Cautiously Optimistic

EDITOR’S NOTE: The following is a rough transcript of comments made by Jill Finn during the Oct. 12, 2015 broadcast of Talk Ten Tuesdays.

The HATA (Healthcare Administrative Technology Association) practice management members as well as our members from organizations like the AMA (American Medical Association) and UnitedHealthcare have been touching base through email almost daily since Oct. 1, and we’re meeting weekly with Denesecia Green, who is the CMS (Centers for Medicare & Medicaid Services) ICD-10 Lead, to exchange issues and find solutions.

From a support perspective, across the board for HATA members, call volumes for our service departments have been lower than anticipated, some as much as 25 percent lower than we had projected, so that’s been great news for our employees who were expecting to be taking calls around the clock.  

We’ve also seen a change in call types, whereas a few weeks ago the majority of calls were around setup and configuration and now we’re seeing a shift into more coding and claims-related questions.

From a provider readiness standpoint, what we’ve seen so far is that a large percentage of our clients are submitting ICD-10 claims, and while that number is increasing on a daily basis, there are still some stragglers. Some are still working through their ICD-9 backlogs, some have processes to submit in batches on a certain cadence (so this could be weekly, bi-weekly, etc.), some are holding onto their claims to double and triple-check their coding, and a few are just letting the dust settle before they submit. What it comes down to is that there’s still a bit of anxiety about getting paid. 

While we hope to not have to utilize this often, one thing that all of our HATA members have done for our providers is build contingency plans into the software to allow offices to toggle between ICD-9 and ICD-10 codes in the event that a payer isn’t ready to accept ICD-10 codes.

As far as ICD-10 claims submission goes, we’re not seeing any glaring issues, which tells us that those years of preparation for our software, staff, and customers has been successful. We’re receiving positive feedback from customers on our readiness, and for the most part, the rejections and exclusions we’re seeing are typical. For the few that aren’t, our clearinghouses and payers have been very responsive. 

Mid-last week, we all started seeing payments come back. AdvancedMD, one of our large billing services, reported that they’ve seen thousands of payments come through. They had payments from Medico in 22 states and insurances like UHC, Tricare, BCBS, Aetna, and Cigna have released or approved the first set of the payments across the states for most of the specialties they work with. 

Medicare has also acknowledged the claims and is currently working through processing, so we should have a solid update on those claims. We’re all patiently waiting to see what the next week or two of adjudication brings and will continue monitoring denials, rejections, exclusions, and other trends.

We’re also keeping a close eye on the few issues that have come out like medical necessity and ensuring that LCD and NCD policies contain accurate information for ICD-10. Having this collaboration across the industry and within our member group has allowed us to keep our services departments and our customers informed and aware of potential impacts.

HATA practice management vendors are experiencing a surprisingly smooth transition for claim setup and submission, and we’re staying cautiously optimistic about a positive claims adjudication experience for our customers in the weeks ahead.

About the Author

Jill Finn is the Information Technology Release Manager at ADP AdvancedMD.  She has more than 10 years of experience in the healthcare IT industry, leading teams to success across a diverse range of products, including practice management, revenue cycle management and human capital management.  She is a board member of the Healthcare Administrative Technology Association (HATA) and served on the Advisory Committee for the Practice Management System Accreditation Program (PMSAP) led by EHNAC and WEDI.

Contact the Author 

Comment on this Article

Coder Productivity Key as Advent of ICD-10 Unfolds

We have been living with ICD-10 for nearly 30 days now, and I think things are going fairly well. What I would like to discuss is coder productivity and accuracy in the hospital setting.  

I was a bit concerned when I was told that offshore coders are only coding 12 inpatient records per day. That is very low in comparison to the typical U.S. productivity standard of 24 per day. Why is it so low? 

I would assume that lack of preparation for ICD-10 had a big impact on the significant drop in productivity. But I would be anxious right now if I had outsourced my coding offshore. It appears that many internationally located coders are overworked, putting in much overtime (which our coders are as well). Also, many are new, inexperienced coders, and they are selecting more unspecified codes, not reading the documentation, and generating significant error rates. In some cases, in an attempt to meet productivity standards, the offshore coders are not capturing the entire clinical picture in the coding, which could have a significant impact to your bottom line. 

In many cases, it does not appear that dual coding took place among many of these offshore companies, which puts them at a huge disadvantage. It is important that if a hospital is using offshore coders, due diligence is performed in finding out if the coders were properly trained and how quality is measured. Productivity was expected to fall by 50 percent or more domestically, but with our contract coders in place, in many hospitals it is hard to measure productivity right now.  

But I have been reviewing individual coder productivity in the U.S., and I am not seeing more than a 5 to 35 percent drop in productivity so far depending on size of organization.

Some of the issues in productivity reduction could be due to issues such as different search terms in the encoder, the need to reference the ICD-10 code book, and anxiety about selecting the right code, especially with the injury codes. Inpatient coders have not had to use code books in years, so that has been challenging as well. Hospitals that have a coding hotline or an experienced ICD-10 coder or consultant available to assist will help keep productivity up.

Most hospitals are going to use contract coders until the end of the year, so we will actually have the opportunity to perform a more accurate measurement of productivity in 2015.  

It is important to first have policies in place to ensure compliance, but also to standardize and set expectations for each coder.  

The first item to consider is accuracy. Most of my clients have set the bar high, between 95-98 percent accuracy. The Office of Inspector General (OIG) standard is 95, but the goal to strive for, of course, is 100 percent. Ninety percent is considered acceptable or commendable, but anything that falls below 90-percent accuracy should be addressed immediately.

One of my clients has a policy that anything below 90 percent results in the coder being placed on probation. The standard for accuracy is reported in three areas:

  • Needs Improvement: 80-89 percent
  • Commendable: 90-97 percent
  • Exceptional: 98-100 percent

I think this is an excellent policy to quantify accuracy expectations.

It is also important to begin to think about analyzing productivity and modifying or changing your productivity goals. I have one client that bases their productivity on a point system and others whose benchmarks are based on how many charts are coded based on type – and quantified by minutes, hours, and date.

Here is the average standard that most small to medium-size hospitals have set: 

Average hospital productivity standard prior to Oct. 1, 2015. 

Type of Claim

Records Per Hour

Records Per Day

Average Time per record




20 minutes

Outpatient (ambulatory, interventional, surgery, and procedures)



12-15 minutes

Emergency department



4-5 minutes

Ancillary services including testing and lab and radiology



2-3 minutes

I have one client that has broken down services based on type of coding, such as IV therapy, surgeries by type, and GI versus interventional, for example. Keep in mind that the more complex the coding, the more time it will take. 

One suggestion to measure where your productivity is now is to look at your productivity by coder and/or type of coding from October 2014 through October 2015 to identify if any productivity loss exists, and if so, the percentage of loss. This should be a report that you run each month for the next 12 months, looking to make strides to get productivity back to what it was in 2014. 

Many hospitals will not use the data from March to Sept. 30 to analyze productivity if they were dual coding during this period because production would be reduced in such an environment.

Stay on top of productivity and accuracy, and if productivity significantly drops, find out why and quickly make corrections to avoid a disaster. 

When monitoring accuracy, it is a good idea to begin auditing inpatient coders now to correct errors and get them on the right track. You should continue to audit and monitor each quarter to ensure accuracy in the coding.

Productivity and accuracy is important to ensure proper payment and getting claims paid quickly. 

About the Author 

Deborah Grider has 32 years of industry experience and is a recognized national speaker, consultant and American Medical Association author who has been working with ICD-10 since 1990 and is the author of Preparing for ICD-10, Making the Transition Manageable, Principles of ICD-10, and the ICD-10 Workbook, among many other publications for the AMA. She is considered an ICD-10 implementation expert and has been helping hospital systems and physician practices get ready for ICD-10 since 2009. She is a healthcare consultant with Karen Zupko & Associates. She is the current president of the Indiana Health Information Association. 

Contact the Author

Comment on this Article

ICD-10 and the Casualty Market: Readiness is Medium to High

The implementation of ICD-10 for auto casualty has been dependent on what happens with the Centers for Medicare & Medicaid Services (CMS) and the rest of the healthcare industry. 

The only states that have a dependency are states such as Pennsylvania and Florida, which use the Medicare rates for payment in auto claims. Pennsylvania adopted Medicare for no-fault payments in 1990 at 10 percent above Medicare and Florida adopted Medicare rates at two times those of Medicare rates in 2013 – with the stipulation that if the 2007 Medicare schedule pays more, it will use it. How is that for an application for a fee schedule? So these states are really being forced into using the CMS payment schemas for payment of claims.

The only other state that has been entertaining the use of a fee schedule using Medicare rates has been Michigan, but for payment of first-party claims only. Michigan also has toyed with the idea of using provider networks and the state’s workers’ compensation fee schedule as well. Michigan has been a state that has had proposals on changing things up for the last 10 years or so, and to this point nothing has passed. 

Implementing ICD-10 in auto casualty is similar to implementing it for healthcare insurers. Even though these payers are non-covered entities under HIPAA, the plan for the majority of casualty payers has been to implement ICD-10 in order to be consistent. The fact that the providers are covered entities and will be sending information to the payers (and receiving information and payments electronically) is also an incentive.

It is our expectation that readiness is medium to high in the casualty market, yet it ultimately will be up to the insurers on what is accepted for payment in the end. Preliminary statistics after the first two weeks of submitted bills since Oct. 1, 2015 are demonstrating nearly 65 percent of providers are using ICD-10 in P&C.  While the codes submitted may not be perfect initially, the numbers are promising. It is well-understood that not all claims will be submitted using the proper format, especially among those practices and providers that typically have billed in only P&C and WC in the past. It will be important for carriers in casualty to address whether they will take a stand or allow some sort of self-imposed grace period for themselves. 

It’s also important to understand that further delays for the casualty payer and provider will not be beneficial as it pertains to the adoption of ICD-10. ICD-10 has as many benefits for patients as it does on the side of providers, and for the casualty payer in communicating why a patient is being treated for any given incident. For example, for auto claims it is important to understand pre-existing conditions because separate policies are held by the patients – and carriers cover them only for the related injuries.

In the casualty industry, ICD-10 also can provide valuable information for car manufacturers all over the world, creating consistency in creating safer vehicles. ICD-10 actually allows for the identification of the side of the body injured in an auto accident, or whether a burn received by a patient was from an airbag deployment. Undeniably, ICD-10 will impact products and safety considerations for consumers considerably, not just in automobiles, but beyond.

For injuries, the most important aspect of ICD-10 (other than being descriptive) is the encounter codes. Having the ability to know whether trauma codes mark a new encounter, follow-up care, or sequela is hugely valuable for payment of auto and workers’ compensation claims. This distinction also may provide more insight into the severity of an injury and potential treatments that are appropriate due to the specificity of the codes. The additional information will cause efficiency gains for the insurer and provider because less clarification and back-and-forth communication will be required.

We have an expectation that there will be changes even after implementation, representing matters that our compliance departments are tracking. The inevitable last-minute changes will happen, and carriers may change their implementation plans on how and what to implement, but the end game will ultimately be the same. 

ICD-10 will finally be used consistently throughout the U.S. to support portability of medical information, even in smaller payment industries such as auto casualty.

About the Author

Michele Hibbert-Iacobacci, CMCO, CCS-P, is vice president of information management and support at Mitchell International, Inc., Auto Casualty Solutions. Iacobacci’s responsibilities in senior leadership include managing health Information, litigation support, industry consultation, regulatory compliance, managed care, and professional services. Iacobacci is a Certified Clinical Coding Specialist (CCS-P), Certified Medical Compliance Officer (CMCO), a Fellow in CLM, and a member of CHIA and AHIMA.

Contact the Author

Comment on this Article