him briefings - enjoinvolume 33 issue no. 2 february 2018 him briefings clinical documentation...

February 2018Volume 33 Issue No. 2

HIM BriefingsClinical documentation improvement

Understanding risk adjustment and compliance for outpatient CDIThere’s no doubt about it: Outpatient CDI programs are growing. In the 2017 HIM Briefings coding productivity benchmarking survey, 13.6% of respondents said they have CDI staff working on inpatient and outpa-tient records, and 8.6% said they are planning to expand to outpatient in the next 12 months. According to the Association of Clinical Documen-tation Improvement Specialists’ (ACDIS) 2017 CDI Industry Survey, more than 30% of respondents plan to expand their CDI program to some type of outpatient service in the near future.

One of the main drivers of the expansion is changing reimbursement models. Reimbursement is increasingly tied to quality, patient outcomes, and the particular health risks facing a patient or patient population. Risk-adjusted and value-based reimbursement models, such as the Quality Payment Program (QPP), Medicare Advantage, or the Hospital Value-Based Purchasing Program, are expanding and impacting reimbursement across the industry. Generally, these programs adjust for risk by calculat-ing risk scores based on Hierarchical Condition Categories (HCC). CDI specialists can make a sizable impact by bringing their documentation expertise and clinical knowledge to bear on outpatient records.

However, organizations and CDI specialists must have a thorough under-standing of how regulations and guidelines impact risk adjustment in the outpatient setting. A misinterpretation can easily lead to inadvertent upcoding—and that can lead to costly audits, settlements, and accusations of fraud. As payment models shift from volume to value, documentation to support risk scores will come under increasing scrutiny.

A new frontier

Traditional fee-for-service reimbursement models, in which the volume of patients impacts reimbursement, are being supplanted by value-based, risk-adjusted models. These reimbursement models place a greater emphasis on the overall health of patients and how much work actually goes into caring for them. A patient with complications/comorbidities (CC) or major complications/comorbidities (MCC) will likely need more intensive care and monitoring—during inpatient acute care episodes and

How to avoid coding issues during EHR implementations

To avoid coding issues during EHR implementation and ensure discharged-not-final-coded is not adversely impacted, dedicated HIM focus and detailed project planning are paramount.

The role of patient access services in denial prevention

This article discusses some of the common issues in patient access ser-vices that impact denials, as well as methods for addressing these issues.

Computer-assisted coding: New techniques to improve accuracy

In advance of ICD-10-CM/PCS, many institutions implemented computer-assisted coding hoping to mitigate the anticipated productivity losses, but research has confirmed suspicions that there is an inverse relationship between coding productivity and accuracy.

A Minute for the Medical Staff

James S. Kennedy, MD, CCS, CDIP, CCDS, discusses how documentation and ICD-10-CM/PCS affect readmis-sion rates.

P5

P7

P9

INSIDE THIS ISSUE

P12

https://revenuecycleadvisor.com/news-analysis/icd-10-productivity-holds-steady

https://acdis.org/articles/outpatient-cdi-getting-and-running

HCPRO.COM © 2018 HCPro, an H3.Group division of Simplify Compliance LLC. For permission to reproduce part or all of this newsletter for external distribution or use in educational packets, contact the Copyright Clearance Center at copyright.com or 978-750-8400.

2 |HIM Briefings February 2018

HIM Briefings (ISSN: 1052-4924 [print]; 1937-7347 [online]) is published monthly by HCPro, an H3 Group division of Simplify Compliance LLC. Subscription rate: Subscription is an exclusive benefit for Platinum members of HCPro’s Revenue Cycle Advisor. Platinum membership rate: $895/year. Back issues are available at $25 each. • HIM Briefings, 35 Village Road, Suite 200, Middleton, MA 01949. • Copyright © 2018 HCPro. All rights reserved. Printed in the USA. Except where specifically encouraged, no part of this publication may be reproduced, in any form or by any means, without prior written consent of HCPro or the Copyright Clearance Center at 978-750-8400. Please notify us immediately if you have received an unauthorized copy. • For editorial comments or questions, call 781-639-1872

or fax 781-639-7857. For renewal or subscription information, call customer service at 800-650-6787, fax 800-785-9212, or email:[email protected]. • Visit our website at www.hcpro.com. • Occasionally, we make our subscriber list available to selected companies/vendors. If you do not wish to be included on this mailing list, please write to the marketing department at the address above. • Opinions expressed are not necessarily those of HIM Briefings. Mention of products and services does not constitute endorsement. Advice given is general, and readers should consult professional counsel for specific legal, ethical, or clinical questions.

This document contains privileged, copyrighted information. If you have not purchased it or are not otherwise entitled to it by agreement with HCPro, an H3.Group division of Simplify Compliance LLC., any use, disclosure, forwarding, copying, or other communication of the contents is prohibited without permission.

justify resource use and truly reflect outcomes?”

It’s natural that organizations and CDI specialists would attempt to frame outpatient CDI and risk adjustment in terms they’re familiar with. For example, a CDI specialist in the inpatient setting who is review-ing a chart for a patient with pneumonia might ask what specific type of pneumonia the patient was diagnosed with. This focus on acute disease manifesta-tions and severity as defined in the APR-DRG model can be beneficial in the inpatient setting, Fee says. However, CDI specialists must take a broader approach in the outpatient setting.

“Certainly, those acute conditions are impactful, especially in HCCs, but there’s a large portion where if we’re only focusing on that philosophy of acute disease manifestation, we’re going to miss out on risk adjust-ment across the continuum,” Fee says.

Chronic diseases are weighted heavily in risk scores, but ensuring they’re documented appropriately and regularly can be a challenge. Providers are generally focused on what they’re treating a patient for on that particular visit, Fee says. The patient may receive treatment for influenza but also have a chronic condi-

during regular, ongoing primary care. Risk scores, based on HCCs, are meant to capture a level of detail that explains the greater amount of resources needed to care for complex patients.

HCCs aren’t new to healthcare. However, in programs such as Medicare Advantage, the commercial payer handles risk score calculations based on HCCs. As a result, provider organizations might not have a solid grasp of risk adjustment. For more on HCCs, see the March 2017 and November 2017 issues of HIMB.

In addition, CDI specialists who got their start on the inpatient side will be accustomed to reviewing charts for documentation and coding concerns that carry more weight in that setting, such as Diagnosis-Related Group (DRG) assignment. Making the leap to outpa-tient might entail a steep learning curve.

“CDI grew up in fee-for-service where most of the opportunity was in the inpatient setting,” says James P. Fee, MD, CCS, CCDS, CEO of Enjoin in Eads, Tennessee, and a hospitalist at Our Lady of the Lake Regional Medical Center in Baton Rouge, Louisiana. “Now we have to extend across the continuum. How do we better define outpatient severity and complexity to

EDITORIAL ADVISORY BOARD

Nicole [email protected]

James H. Braden, RHIA, MBASenior ConsultantThe Advisory Board Company Nashville, Tennessee

Ben Burton, JD, MBA, RHIA, CHPS, CHC, CRCConsultant First Class Solutions, Inc. Maryland Heights, Missouri

Jean S. Clark, RHIA, CSHAHIM and Accreditation ConsultantChicago, Illinois

Rose T. Dunn, MBA, RHIA, CPA, FACHE, FHFMAChief Operating OfficerFirst Class Solutions, Inc. Maryland Heights, Missouri

James P. Fee, MD, CCS, CCDSVice President, Enjoin Hospitalist, Our Lady of the Lake Regional Medical Center Baton Rouge, LA

Darice M. Grzybowski, MA, RHIA, FAHIMAPresidentHIMentors, LLC Westchester, Illinois www.himentors.com

Anita Hazelwood, RHIA, FAHIMAProfessor, HIMUniversity of Louisiana at Lafayette Lafayette, Louisiana

Elaine Lips, RHIA Vice President HIM & CDI Services Xtend Healthcare Los Angeles, California [email protected]

Monica Pappas, RHIAPresidentMPA Consulting, Inc. Long Beach, California

Glennda Rinker, MA, RHIADirector of HIM and HIPAA Privacy OfficerWoodland Heights Medical Center Lufkin, Texas

Chris Simons, MS, RHIAHIM Director and Privacy OfficerMaineGeneral Health Augusta, Maine

http://hcpro.com

https://revenuecycleadvisor.com/archives?pub=278&volume=34&issue=3

https://revenuecycleadvisor.com/archives?pub=278&volume=34&issue=11

HCPRO.COM© 2018 HCPro, an H3.Group division of Simplify Compliance LLC. For permission to reproduce part or all of this newsletter for external distribution or use in educational packets, contact the Copyright Clearance Center at copyright.com or 978-750-8400.

HIM Briefings | 3February 2018

tion such as multiple sclerosis (MS). If the MS isn’t appropriately documented annually, or at a particular visit if it’s discussed, the omission could skew the patient’s risk score.

“A lot of people will target those conditions that are MCCs or CCs, which is not going to have an impact in an outpa-tient setting,” Fee says. “On the contrary, if you’re looking for chronic disease, that can drive those risk-adjusted performance measures within the inpatient setting.”

It might appear simpler—and more comfortable for the organization and CDI staff—to focus outpatient CDI efforts on departments more closely tied to the hospital such as an emergency department or ambulatory surgery center, says Sonia Trepina, MPA, director of ambulatory CDI services for Enjoin in Eads, Tennes-see. However, depending on the individual organiza-tion, the best opportunity for outpatient CDI might lie well outside of that comfort zone.

“We have found that if you’re entering into risk, the more risk you’re entering into as an organization—whether that’s through Medicare Advantage programs, becoming a payer, becoming part of an ACO—there’s opportunity in the physician office setting,” Trepina says.

A physician office might also represent the biggest challenge to an outpatient CDI program. Physicians were traditionally reimbursed based on evaluation and management (E/M) services. Complete documentation for the ICD-10 code diagnostic component of E/M services requires a maximum of only four diagnoses, Fee says. Entering into risk-adjusted models requires physicians to make a major shift in their documenta-tion. A complex patient will require more than four diagnoses—including past surgeries or conditions such as a myocardial infarction.

To make things even more challenging, many patients are seen only a few times or even just once each year. That drastically reduces the opportunities to capture data, making it even more important that providers document thoroughly for each encounter. However, providers shouldn’t view documentation as simply an administrative burden. The more information a provider collects from a patient, the better the care will

be, Fee points out. Each piece of data helps the pro-vider create a larger picture of the patient’s overall health and allows tracking of health trends for indi-viduals and populations.

Nevertheless, it’s no surprise that physicians already feel overwhelmed by documentation requirements. Primary care physicians (PCP) in particular are key to document-ing chronic conditions captured in HCCs. A chronic condition only needs to be documented once a year, but organizations should keep in mind that PCPs typically have a heavy workload. Capturing all chronic conditions can turn out to be a taller order than it initially appears.

“The PCPs become really important to risk adjust-ment,” Trepina says. “But just the sheer volume of patient loads and what they need to capture, even though they only need to capture it once a year for HCCs, it can be overwhelming.”

Getting up to speed

Misconceptions about risk adjustment are common, Fee says. Some organizations might struggle to get a handle on HCCs, while others may assume that HCCs are the only factor that influences risk and quality calculations and then fail to hold other aspects of documentation to the appropriate standard.

“For ambulatory CDI, the focus on documentation compliance, accuracy, and completeness is important, but reporting is critical and often overlooked,” Fee says.

Organizations must keep in mind that risk adjustment also impacts an organization’s quality measures. This point is often overlooked, Trepina points out, leaving an organization vulnerable on multiple fronts.

“Not a lot of people realize that when you capture certain diagnoses and certain codes in the ambulatory setting, that impacts your quality measure, for exam-ple, in your hospital setting,” she says. “The risk adjustment piece is so critical.”

Compliance pitfalls

Putting the focus on risk adjustment means organiza-tions need to hold documentation and staff to a higher standard. Even a suggestion of inappropriate coding

http://hcpro.com



and documentation that makes patients appear sicker than they truly are can land organizations, providers, and other staff in hot water.

As organizations move to risk-adjusted models and train staff, they must stress the role of compliance. The organization should also take a look at its EHR system and how users interact with and are influenced by the system’s interface. A particular EHR can, even inad-vertently, guide the information a provider enters. In turn, that impacts coding.

“The EMR and the provider choosing the diagnosis code becomes the first step in the coding process,” Trepina says. “Making sure that providers know that what they’re clicking is the right thing is important.”

Some EHRs are set up to drive HCC maximization and optimization. Vendors and some other stakehold-ers might see this as a feature that can help remind busy providers, but it can easily become a fast track to noncompliant documentation. Some EHRs that are set up to identify HCCs may pose what can be construed as leading documentation options to providers.

“A physician might forget that when you choose a diagnosis, it has to be supported in the documenta-tion,” Fee says. “Something has to be done for that condition, and if it’s not and you’re reporting that because it’s an HCC, then there’s a compliance risk.”

CDI specialists can help head off those compliance risks, Trepina says. CDI reviews shouldn’t be limited to records that might have an opportunity to capture more HCCs. Instead, CDI specialists should also take a look at records where the risk score is fairly high and ensure that the documentation supports that score, she says. In either case, CDI specialists should have the resources to offer education to providers on appropri-ate capture and documentation of HCCs.

“The provider doesn’t need to write paragraphs,” she says. “It’s not the length of the note: It’s the choice of words to accurately describe the situation.”

The challenge is twofold: to capture a risk score that reflects the complexity of a patient and to ensure that the supporting documentation is complete and compli-

ant. In service of this aim, organizations should build infrastructure that continually supports and reinforces providers, Fee says.

“Providers have a lot on their plate managing care for the patient,” he says. “So, having an infrastructure built to keep them up-to-date, to reinforce what they’re doing and why they’re doing it and not lose sight of it, will help maintain a healthy program long term and will avoid having OIG knocking at your door.”

Working together

Collaboration is key to a successful outpatient CDI program, Fee says. CDI grew up on the inpatient, acute care end of the spectrum and generally had little contact with or awareness of concerns that impact outpatient documentation. When a CDI program expands, the CDI specialist may not know whom to contact to help the outpatient branch of the program thrive. CDI specialists should reach out to physician offices and other outpatient facilities to establish communication and connections.

“Once you get involved, then you get to understand what’s happening on that side of the world and where you can help,” Fee says. “Get involved in as many meetings as possible where coding and documentation is impacted.”

The patient’s health and well-being must be the focus, Trepina points out. Risk adjustment is ultimately about the patient. Organizations should keep that perspective in mind, and CDI specialists should act as the advocate for translating that principle into documentation. Correct, complete documentation leads to correct code assignment, and that in turn helps providers, payers, and organizations understand how to care for patients and allocate resources.

Questions? Comments? Ideas?

Contact Associate Editor Nicole Votta at [email protected]

http://hcpro.com

mailto:?subject=



The implementation of an EHR is a multifaceted, comprehensive project for healthcare organizations. Though every department and function is affected in some way, the impact on medical record coding may be the most profound and complicated, with some organi-zations reporting a 20% decrease in coding productiv-ity post-implementation.

To avoid coding issues during EHR implementation and ensure discharged-not-final-coded (DNFC) is not adversely impacted, dedicated HIM focus and detailed project planning are paramount. Following are five recommendations to protect your coding process and support a successful EHR implementation.

Hire an HIM project manager

Designate a project manager (PM) within HIM to be the dedicated point person for the project.

This is especially important if coding is centralized, or will be centralized, as part of the EHR implementa-tion. Ideally this person will be a member of the HIM management team and have an overall understanding of the coding process. The PM should oversee and participate in the following activities:

• Assess coding practices throughout the organiza-tion. Determine who is responsible for existing pro-cesses and what the expectations will be moving forward.

• Identify areas responsible for specific types of coding such as interventional radiology, and whether that structure will continue after go-live.

• Determine who is authorized to make changes to coding sets in the system.

• Ascertain how charges are entered on bills. Iden-tify all chargemasters in use and verify the infor-mation within each.

• Review all coding policies and procedures. If there are departments without documented poli-cies and procedures, compile the information and make it available online prior to go-live.

• Understand how information is feeding into the

current coding system(s) and who has access to re-ports and queues. Create a list identifying a point person for each area entering information or gen-erating a report.

• Meet with each point of contact to determine how frequently each report is run and how meaningful the data is. How is the information used? Will it be needed going forward?

• Find out how the data will look on reports gener-ated from the new system. If there are differences, cross-reference all items so that users can quickly navigate the new report.

• Determine if the facility will have access to his-torical data; if so, how will it be accessed?

• How can future report queries based on historical data be made in the new system?

• Does the new system have the ability to auto-run routine reports? If so, ensure that all policies and procedures spell out the deadlines by which infor-mation must be entered.

• Specify how each clinical department or service line documents professional fees and identify poli-cies that guide the process.

• Review current coding queries and templates. Can they be duplicated in the new EHR? Are they con-sidered part of the legal record? Will any changes need to be approved by compliance? If so, what is the time frame to accomplish that process?

Engage a physician liaison

Designate at least one physician liaison to interact with coding and CDI leadership throughout the EHR project. The ideal candidate remains closely involved in all discussions related to documentation require-ments for providers, including how to handle documen-tation deficiencies and coding queries in the new EHR. The liaison should also be involved in determining how and when physician documentation training will occur. If the organization is a teaching hospital, different training approaches and modules may be required for new residents versus staff physicians.

How to avoid coding issues during EHR implementationsby Charlene M. Laxson, CCS

http://hcpro.com



Participate in go-live planning

Provider organizations lean toward two types of go-live strategies. One is a big-bang approach, where all modules are implemented at the same time. The other is a phased-in approach, where one or two modules at a time are implemented over several months or years. Coding leadership should understand which approach will be used and share coding work-flow concerns with the EHR project team during initial go-live discussions.

Coding and other HIM leaders also provide revenue cycle insights from a holistic, big-picture perspective—valuable input that is often overlooked by IT teams and EHR vendors. Therefore, their participation is critical to maintain consistent cash flow during imple-mentation and conversion.

For example, important decisions around EHR testing—how, when, who—can adversely impact clinical documen-tation and coding workflow. End-to-end testing of all processes is essential, and the HIM project manager should ensure information flows correctly to all coding queues. Likewise, important billing questions should be discussed jointly with IT, revenue cycle, and coding teams prior to EHR go-live, such as verifying that all bills have the correct charges attached to each code.

Another important multi-stakeholder discussion involves patients admitted prior to go-live and discharged after-ward. Discuss and address the following questions to avert a negative impact on coding workflow:

• Will patients be discharged from the old system and admitted into the new system?

• How and when will bills for the stay be processed?

• Will patient information such as test results, prog-ress and nursing notes, and drug orders be copied into the “new” admission in the new system?

• What will be involved with any transfer of infor-mation process, and who will be responsible for ensuring timely completion?

• How will the organization ensure cases are cod-ed and bills dropped in a timely manner after go-live? Will additional support be needed in CDI, coding, or billing?

Plan well in advance

Ideally, the HIM and coding department is involved in all steps of an EHR implementation and go-live planning. Time frames and milestones should be known well in advance of any potential workflow impact. In addition, notify outsourced coding vendors at least six months prior to the scheduled go-live. Contract with them for additional services and support to avoid backlogs and maintain day-to-day operations while internal staff trains, tests, and ramps up skills using the new EHR.

The coding vendor will require the following project details to ensure proper contract execution:

• Implementation timeline

– Training dates for your coders as well as for con-tract coders

– Remote coders’ knowledge of specific computer platforms needed with new software

• Duration of training

– How many days of training is each coder expect-ed to attend, and how many hours per day?

– Will training be phased over several days/weeks, or will all staff be trained at the same time?

• Training locations

– Will training be conducted on-site or remotely?

• Training details

– Will coders perform system testing? If so, will they need to be back-filled, and for how long?

– Will you require additional staffing during coder training?

– How will the coding system go-live be executed? Will the contract coders work only in the old sys-tem, or will they divide their time between both systems?

Addressing these points in advance alleviates many potential coding workflow issues with EHR implementations.

Conduct thorough testing

Despite the best planning, coding hiccups will happen. We know of instances where it was discovered after go-live that coding queues weren’t populating cor-

http://hcpro.com



rectly. Others have had issues with correct charges not flowing to the bill via their chargemaster. These issues take time to troubleshoot and resolve, ultimately delaying the coding and billing process.

We recommend organizations take time to test every single function prior to go-live to ensure all information is flowing appropriately. Consider the case of the charge-master issue mentioned above. The organization had to go back and conduct a line-by-line check of each depart-ment’s ICD-10, CPT®, and revenue codes to identify the root cause of the issue. The go-live process could have been much smoother had the organization thoroughly tested the chargemaster prior to implementation.

Testing to this level of detail is time-consuming, but the benefits are worth the effort. Even if every item is checked and double-checked, adverse coding impacts may occur. Best practice is to have processes in place to

track all issues identified during go-live and ensure all issues are resolved—and have a backup plan should hiccups occur.

Coding backlogs impact cash flow and financial health. Protecting coding productivity and workflow during EHR implementations is a critical executive concern.

EDITOR’S NOTELaxson is a coding manager with KIWI-TEK, a coding consultant organi-zation. Laxson enjoys tackling hospital revenue cycle challenges from coast to coast with hospitals of all sizes and coders with all levels and areas of expertise. Prior to her role with KIWI-TEK, Laxson worked with Scott and White Medical Center in Temple, Texas, in outpatient coding management and coding and reimbursement analysis. Laxson is a Cer-tified Coding Specialist (CCS) who has worked in the HIM field for 15 years. She can be reached at [email protected]. Opinions ex-pressed are those of the author and do not necessarily represent HCPro, ACDIS, or any of its subsidiaries.

The role of patient access services in denial preventionEducating patient access staff plays a key role in keep-ing denial rates low.

Often, when a hospital is doing well, it retains educa-tors and quality assurance (QA) managers to help with training, quality review, updating, and implementing new systems. When a hospital is not doing well, how-ever, these positions tend to be eliminated, and the responsibilities are assigned to departmental manage-ment. The challenge is that managers and supervisors have many obligations and deadlines to meet, and it is difficult to keep the staff updated and trained in their positions and to process daily QA on top of existing duties. If a hospital wants to reduce denials and ensure revenue integrity, therefore, it should retain an educa-tor or QA manager. You cannot afford not to have one.

The educator or QA manager is the main communica-tor between patient accounts, registration, information technology, case management, health information management, and managed care contracting. This person can serve as the main trainer and can keep patient access representatives updated in all areas of

the registration department.

This article discusses some of the common issues in patient access services that impact denials, as well as methods for addressing these issues.

Education registration services

Train all patient access staff on the registration process. They should understand types of insurance (government, commercial, and third-party administrators), pre-certifi-cation, authorizations, and verification of coverage.

Educate staff on denials as well, whether via daily QA; monthly superuser meetings; implementing new systems or updates to registration software; or moni-toring copayments, coinsurances, and deductibles, and eligibility/verification.

Ensure that the insurance and coverage rules are updated in the system weekly or monthly. Doing so allows tracking of regulatory changes and will keep staff up-to-date on the latest rules that affect the registration process.

http://hcpro.com

mailto:[email protected]



Registration denials

It is easy for registration staff on the front end to make errors, but there are many errors made by other departments that contribute to costly denials. Although each department must accept responsibility for errors originating with it and take action to improve, all departments in the revenue cycle ulti-mately must work together to ensure success.

It is not uncommon for a denial to originate with the registration or patient access department. Denials can result in missed or lost revenue, so it is critical for revenue cycle professionals to determine the root cause of denials at their facility. That often means starting with the actions of the first staff members to encounter the patient.

There are a variety of ways in which patient information collection and entry by patient access staff can lead to an error in the medical record that later results in a denial.

Insurance information

Incorrect or incomplete documentation of the patient’s insurance information can lead to problems with billing for patient care.

Although each hospital must understand what causes denials in its specific case, there are many kinds of denials that are common among hospitals. Patient access managers must be aware of those causes that originate within their department and should imple-ment methods of addressing these problems.

Demographic entry errors can be avoided by paying close attention to the information being entered, reviewing information before it’s submitted, and ensuring that complete information is entered. The following is a list of some common demographic data entry errors, causes, and suggested resolutions.

• Patient’s address was incorrectly entered. A simple mistake such as transposing digits can cause the claim to be denied. To prevent such mistakes, pa-tient access representatives should be careful typ-ists and double-check all information they enter. In addition, staff should ensure that the address on file is the patient’s most current address. They can do so by checking the address on file against the in-formation the patient provides, such as a driver’s

license, or by asking the patient to confirm the ad-dress verbally. Rather than reading the address on file and asking the patient to confirm that it is cor-rect, ask the patient to state his or her address.

• Incorrect date of birth entered. A date of birth that does not match that on file with the insurer can lead to a denial. Patient access representatives should be careful to enter the correct date.

• Patient’s name is incorrect. When entering insur-ance information or checking for eligibility, the patient’s name must match the name on the insur-ance card exactly. For example, if the patient’s le-gal name is John Adam Murphy but his Medicare card reads Jack A. Murphy, enter the latter.

• Insurance ID number entered incorrectly. Insur-ance ID numbers generally mix alpha and numeric characters. It can be very easy to transpose or mis-take characters, particularly during peak hours or if patient registration is short-staffed. For example, an insurance card that reads YRI234M6488 could be mistyped as YR1234M6488. Or an insurance card that reads 1123468902 could be incorrectly entered as L123468902. Patient access representa-tives should carefully check insurance ID numbers to ensure that they enter them correctly.

All departments need to be accountable for their errors. Errors that result in denials or that impact revenue integrity should be explained and appropriate education provided. If an error continues, consider a process improvement plan for a department or individual.

Recommendations to reduce denials

It is highly recommended to retain an education coordi-nator or quality assurance manager for patient registra-tion. All staff should take regular refresher classes. If there is a significant change or update, schedule a class to push out education as soon as possible.

Provide training on new systems before go-live. Staff will need time to learn the new system, and providing training in advance will help identify bugs or other issues that should be addressed before the system is fully implemented.

Staff should have regular and update-dependent classes on the following key topics:

http://hcpro.com



• Medicare, including Medicare Secondary Payer

• Other types of government insurance, such as TRICARE

• Commercial insurances, including out-of-network insurances commonly seen in your facility

• Reasons for denials, including common and re-peat causes

All departments in the revenue cycle should work closely together to reduce denials. Invite all staff to the appropriate meetings. Ultimately, everyone is a stake-holder, and reminding staff of the important role they play in keeping the doors open and providing care can increase commitment and job satisfaction.

Hold bimonthly superuser meetings to update all staff of changes, and hold bimonthly meetings for all revenue cycle departments to discuss denials.

Reducing denials and maintaining revenue integrity is an ongoing process. Monitoring goals and encouraging staff to meet them on a continuous basis should be part of that process. For example, a hospital may train all patient access representatives on point of service (POS) collec-tions in June. In July, POS collections increased by 25%–30%. However, the manager did not continue to encourage staff to maintain POS collections in the follow-ing months; therefore, POS collections decreased again.

Additionally, hospitals may consider requiring patient access representatives to obtain the National Associa-tion of Healthcare Access Management’s Certified Healthcare Access Associate (CHAA) certification within one year of hire.

Summary

Patient access managers must be engaged members of a facility’s denials management team. Simple data entry errors are easy to make, particularly during high-vol-ume hours, but reducing these should be a priority. Patient access representatives must be well-versed in insurance identification, verification, and eligibility. The value of correct, complete information collected when the patient enters the system cannot be overstated.

Patient access staff also play a critical role in the pre-authorization process and ensuring timely delivery of required notices.

EDITOR’S NOTEThis article was adapted from The Complete Patient Access Hand-book, by Marsha L. Sopiecha. Contact Editor Nicole Votta at [email protected] for more information.

Computer-assisted coding: New techniques to improve accuracyby Erica E. Remer, MD, FACEP, CCDS

In advance of ICD-10-CM/PCS, many institutions implemented computer-assisted coding (CAC) hoping to mitigate the anticipated productivity losses, but research has confirmed my suspicions that there is an inverse relationship between coding productivity and accuracy.

Sacrificing accuracy on the front end guarantees resource waste on the back end when appealing denials. Reimbursement and quality metrics suffer with diagnosis removal and MS-DRG downgrade,

being dependent on the codes accurately representing the complexity and severity of illness.

Users describe CAC technology as being “in its infancy,” although I wonder if describing it as the “toddler stage” is more apropos. Companies have been trying to evolve and reprogram the background logic in response to the feedback from early adopters and current utilizers. In this article, I am going to highlight some of the pitfalls of CAC and techniques

http://hcpro.com

https://hcmarketplace.com/the-complete-patient-access-handbook

https://hcmarketplace.com/the-complete-patient-access-handbook





to improve accuracy. Perhaps we can get through the terrible twos together.

Understanding CAC

CAC is to native, manual coding as reading Dickens’ Great Expectations for an English class is to reading the book for pleasure. Having to analyze the theme, character development, setting, symbolism, and language affords you a completely different experi-ence than just enjoying the story. Coding is dependent on understanding the narrative of the patient encoun-ter and being able to coherently translate it into accurate, specific codes.

The best user of CAC first reads the record and understands the sequence of events and how the encounter unfolded. The coder can then can examine each offered code and assess its applicability. In fact, the most difficult part of using CAC is resisting the urge to relinquish the coding to the computer.

Novices and experienced coders alike may be over-whelmed by the large number of autosuggested codes. There do not seem to be readily available statistics on the average number of codes a CAC system offers up, but a super user I interviewed referenced a 12-day stay providing him with 55 codes. I imagine this makes it quite challenging to separate the wheat from the chaff. Additionally, the CAC color-coding can be quite distracting; your eyes are drawn from one highlight to the next, and you might be disinclined to read the un-highlighted snippets between, which could contain important information.

Sometimes, in an attempt to make the electronic health record (EHR) more functional, providers alter the font of their text. Attending physicians bold, italicize, or color their additions to a resident’s or advanced practice practitioner’s note. CAC does not take format-ting into consideration, so the attending physician’s documentation holds no more sway than anyone else’s. This may set up conflicts, or a coder may miss an attending discounting a resident’s diagnosis.

Content versus context

Computers are only as good as their programming. Documentation and coding are based on complex

decision trees, and computers do not think like human beings. The computer may take verbiage of “heart attack” and autosuggest code I25.2 (old myocardial infarction), even if the doctor qualifies it with “last week.” If the coder doesn’t notice that the myocardial infarction occurred within 28 days, a legitimate MCC may be lost. Computers are much better at assessing content than they are at context, but coding often requires command of both.

When I do a comprehensive evaluation of a medical record and find a patient in an inappropriate MS-DRG, one cause may be sign and symptom DRG aversion. If, however, CAC is the culprit, it’s because the system has found a definitive condition that may be manifested by a sign or symptom (e.g., weakness/acute kidney injury [AKI]), and it suppresses the autosuggestion of the sign or symptom. A fact of clinical life is that sometimes patients get admitted for persistent or severe signs or symptoms, and no definitive diagnosis is ascertained after study; these patients get better despite us. If the only diagnosis a coder has for a four-day stay is AKI that resolved on day two, the patient ends up erroneously in the renal failure DRG. DRG 684, Renal Failure without CC/MCC, may have more favorable metrics, but DRG 948, Signs and Symptoms without MCC, would be the correct DRG.

Another example is the symptom of agitation or restless-ness getting suppressed by the accepted CAC suggestion of F32.9 (major depressive disorder, single episode, unspecified) for the verbiage of “history of depression.” Agitation and restlessness are sometimes accompanied by behavioral disturbance in dementia, but the coder needs to see the big picture to tease that out.

Nuance is not a computer’s strong suit. I once reviewed a case where a patient had chronic kidney disease (CKD) with superimposed AKI requiring emergent dialysis. The CAC suggested end-stage renal disease (ESRD) as a diagnosis and excluded the AKI and CKD. This is a tough concept to explain to a machine: Dialysis is usually, but not exclusively, associated with the chronic condition of ESRD.

Linkage is also challenging for CAC. A system can be erratic in finding combination codes or the most

http://hcpro.com



specific codes. One CAC system can link diabetes mellitus (DM) with CKD found as two separate entries, but it can’t make the leap for diabetic gastro-paresis. Is this because the title of the code is “DM with diabetic autonomic (poly)neuropathy?” Whatever the reason, it stumbled, taking the verbiage of “neu-ropathy due to Lyme disease” to the unspecified code instead of A69.22 (other neurologic disorders in Lyme disease).

Finally, if a provider uses dictation software or is a lousy typist, you don’t want unconventional grammar or creative spelling to prevent an accurate representa-tion of the patient’s condition. CAC can’t find what it hasn’t been programmed to register.

Assessing CAC relevance

Computers excel at capacity. Every day I use ICD-10-CM, and every day I find cool, new (to me), useful codes that I didn’t realize existed. CAC has access to the entire set of ICD-10-CM codes in its database, so it is highly possible that it can pick up documentation that the coder might not recognize as having a unique code. This probably maximizes specificity and com-pleteness, although on occasion its relevance may be dubious.

Relevance is the issue I have with superfluous per-sonal history and family history Z codes. Secondary diagnoses are often quoted as having to be evaluated, monitored, or investigated, or as having to increase nursing care or length of stay. The original Federal Register from 1985 also included “conditions that coexist at the time of admission,” but some such conditions have no relevance to the current encounter. Documenting and coding every historical condition a patient has ever had is excessive and can be distract-ing. If a condition does not impact the provider’s care or does not affect the patient in the present, it is not relevant and the coder should not be capturing it, even if the CAC flags it.

CAC’s most basic task is performing word searches, but sometimes it fails even at this. I had an example submitted to me where a search for the word “sepsis” came up empty, but in the same record, the search for documentation support for the suggested code of

A41.9 (sepsis, organism unspecified) produced a consultant note’s discharge list that included sepsis. For technology to be useful, its results must be reliable and reproducible.

Using CAC effectively

The bottom line is that CAC is not infallible, and it is not a panacea. It is a powerful, useful tool to be wielded by coders who understand its idiosyncrasies and limitations and who are confident in their own abilities. This leads to the obvious question, “What is the best approach for using CAC effectively?”

First, the foundational culture must reject sacrificing accuracy for speed. Productivity metrics should not be paramount, or at least not be measured in a vacuum.

Novice coders should have an initial period of time where they code natively. If they use an encoder, they need to understand that it too has shortcomings; the source of truth is the coding book and guidelines. The medical record must be read in its entirety and from beginning to end to appreciate the evolution of the encounter. Only after proving they have the requisite depth of clinical knowledge and discipline should coders be permitted to use computer assistance.

Super users should have a set of examples to train new CAC users on that expose them to the hazards of overreliance on the computer brain. When users run across flawed logic, they should have a system contact person who can give the CAC vendor feedback. If your system hasn’t selected a vendor, I recommend contacting references and investigating whether the vendor is receptive and responsive to user input and suggestions. Submitting feedback and continually getting the same wrong output can drive users a little batty, in addition to resulting in inaccurate coding.

When using CAC, the coder must still maintain the discipline to read the chart in its entirety—even, or perhaps especially, the un-highlighted text. Bear in mind that CAC is suggesting codes for coder valida-tion, not making decisions.

After compiling the set of codes for an encounter, the coder should review them and see if they tell the story.

http://hcpro.com



Documentation and ICD-10-CM/PCS affects readmission rates


Dear Colleagues:

Like it or not, CMS, the Veterans Health Administration, and some states (e.g., Illinois, Texas, Maryland, Florida) measure our care quality based on risk-adjusted readmission rates after inpatient admissions. In fact, up to 3% of our hospital’s Medicare inpatient revenue (used to pay physician subsidies) is at risk if we don’t manage our patients’ readmissions in concert with Medicare’s algorithms. Other payers, such as Anthem Blue Cross Blue Shield, also use Medicare’s readmission measure to create quality-based risk models for their non-Medicare business. Certain Medicaid programs use other algorithms, such as the 3M Potentially Preventable Readmissions measure, that result in an even higher percentage of total revenue lost.

Just how much money is at risk? Well, you can look up your hospital traditional Medicare performance on the FY2018 Medicare IPPS Impact File under column BB, available at www.cms.gov/Medicare/Medicare-Fee-for-Service-Payment/AcuteInpatientPPS/Downloads/FY2018-CMS-1677-FR-Impact.zip. If you wish to know the dollars lost by your facility, multiply (1.00 – column BB) x Column AJ x Column AR x $8,000 (an average Medicare base rate; use your hospital’s own Medicare base rate if you know it).

A review of Medicare data for FY2013–FY2016 available at https://data.medicare.gov/Hospital-Com-pare/Unplanned-Hospital-Visits-Hospital/632h-zaca suggests these and other readmission penalties could be ameliorated or even eliminated if hospitals work with us to ensure the validity of our outpa-

If they do not accurately represent the sequence of events and severity of illness, something is missing. Sometimes it is a code, and sometimes it is a condition that is not yet codable because the documentation is lacking. This should elicit a query.

There should be periodic auditing of the coding to ensure there has not been CAC-induced degradation of coding quality. Each facility should be thinking about internal audits, external audits, how many charts should be looked at, and how often.

Finally, there must be a feedback loop and education. Coders should be informed of their accuracy and productivity metrics, but this is useless without also telling them how they can improve. Everyone learns

best from directed, case-based examples. I am a strong proponent of disseminating useful information so everyone can learn from everyone else’s mistakes.

There is a reason the technology is called “computer-assisted” coding, and not computer-coding with coder assist. There is no substitute for a competent and vigilant coder. If you use CAC wisely, however, you can have the best of both worlds: increased productiv-ity and improved accuracy.

EDITOR’S NOTEDr. Remer is founder and president of Erica Remer, MD, Inc., Con-sulting Services. Contact her at [email protected]. Opinions expressed are that of the author and do not necessarily represent HCPro, ACDIS, or any of its subsidiaries.

http://hcpro.com

https://data.medicare.gov/Hospital-Compare/Unplanned-Hospital-Visits-Hospital/632h-zaca

https://data.medicare.gov/Hospital-Compare/Unplanned-Hospital-Visits-Hospital/632h-zaca




tient and inpatient clinical documentation applicable to the readmission metric. That, in turn, ensures the ICD-10-CM codes impacting cohort selection and risk adjustment are properly adjudicated for the high-risk patients we took care of that ended up requiring a readmission. If we, as physicians, learn the foundations of readmission measurement and implement some basic principles and workflows for reporting clinically accurate ICD-10-CM/PCS coding that affects these outcomes, we can help our hospitals succeed in their readmission measures, reflect the excellent work we do in caring for our patients, and have the resources to solidify our partnerships with them.

Need to get that O to E right

Most physicians and case managers are not aware that CMS uses ICD-10-CM and ICD-10-PCS codes to determine whether patients qualify for inclusion in the readmission cohort and if their illnesses predict a higher risk of readmission. Once the cohort has been established, our risk-adjusted readmis-sion rates are as follows:

Risk-adjusted Readmission Rate = Predicted (Actual or Observed) Readmission Rate Expected Readmission Rate

If one’s risk-adjusted readmission rate is high, it can be improved by either decreasing the Observed (or Actual) Readmission Rate through aggressive (and expensive) case management or increasing the Expected Readmission Rate by ensuring that the documentation and ICD-10-CM/PCS coding cap-tures all the risks that would have predicted the readmission in the first place. In other words, even if my Observed Readmission Rate remains the same as a numerator, increasing my Expected Readmis-sion Rate as a denominator lowers my risk-adjusted Readmission ratio. The same concept applies in all of CMS’ risk-adjustment methodologies, such as costs, mortality, length of stay, or complication rates.

In illustrating this concept, let’s look at CMS data for pneumonia readmission between FY2013–FY2016, for two fine hospitals, Hospital A and Hospital B. (Editor’s note: Although the names of the organizations have been changed, the examples given are based on actual CMS data.)

Hospital Name Number of Discharges

Excess Readmission Ratio

Predicted Readmission Rate

Expected Readmission Rate

Hospital A 1284 1.1492 18.5 16.1

Hospital B 820 1.0112 18.9 18.7

Note how Hospital A has an Actual Readmission Rate that is lower than that of Hospital B; however, since Hospital B’s inpatient and outpatient documentation and coding of patient comorbidities reflected a higher readmission risk, Hospital B’s readmission score was much better than Hospital A’s. I would say that if Hospital A or any other hospital we know wanted to get out of the penalty box ($4.4 million worth), that hospital would want us to better document and code these outpatient (within 12 months of the index admission) and inpatient diagnoses that impact their Expected Readmission Rate.

Access the following CMS database to run your own analysis for your facilities and those in your locality to determine if your facility’s successes or challenges are a numerator or denominator issue: https://data.medicare.gov/Hospital-Compare/Hospital-Readmissions-Reduction-Program/9n3s-kdb3. Your discussion of this at a readmission prevention steering committee will likely bear fruit.


http://hcpro.com

https://data.medicare.gov/Hospital-Compare/Hospital-Readmissions-Reduction-Program/9n3s-kdb3




When all else fails, read the directions

It would be impossible to explain all of CMS’ readmission methodologies in one article. Please consider reviewing CMS’ readmissions definitions manuals available at http://tinyurl.com/mmvprzy or reading a short summary of a lecture I gave on this subject at the American Case Management Association Leadership Conference in New Orleans, available at www.acmaweb.org/Leadership/2017/17LPA_ Session5_Kennedy_Notes.pdf. Salient points include:

• Strengthen your discharged against medical advice protocols CMS makes it very clear that patients who are discharged against medical advice (AMA) are not included in the readmission cohort since the provider loses control of the patient’s dynamics in re-admission prevention. Of course, we must do all that we can do to prevent the AMA discharge; however, if a patient insists on leaving despite our best efforts, our documentation must be crystal clear that the discharge was against the patient’s best interest and our advice so that the medical re-cords abstractor assigns the correct AMA discharge status to the encounter.

• Clearly document any planned readmissions Most of us do not know that inpatient billing uses UB-04 discharge status codes that report wheth-er we plan to readmit the patient at a further date, which, if communicated, may not be considered as a penalty. Unless we document the planned readmission clearly, the medical records abstractor may not capture these codes. These will likely not affect CMS’ readmission measurements; howev-er, they may affect others. A summary of these are available at www.bcbsks.com/CustomerService/Providers/Publications/institutional/newsletters/2013/09-10_new-patient-discharge-codes-effec-tive-10-1.htm.

• Clearly document all the reasons for inpatient admission on the discharge summary A common misperception is that CMS readmission measures are based on the Medicare-Severity Diagnosis-Related Group. This is incorrect; the readmission measure, for the most part, is based on what ICD-10-CM code is sequenced as the principal diagnosis, defined by ICD-10-CM as the con-dition found after study (emphasis added) to have occasioned the inpatient (emphasis added) admis-sion to the hospital. Examples include:

– A patient is admitted with decompensated heart failure due to rapid atrial fibrillation. If the docu-mentation clearly states that the rapid atrial fibrillation contributed to the heart failure decompen-sation and required treatment along with the heart failure, sequencing the ICD-10-CM code for atrial fibrillation as the principal diagnosis instead of heart failure ICD-10-CM codes removes the case from the heart failure readmission measure.

– A patient is admitted with decompensated heart failure but with a substantial rise and fall in the serum troponin with associated “demand ischemia.” If the documentation at the time of discharge is clear that the rise and fall of the troponin was consistent with a demand myocardial infarction (MI) (not ischemia) and the appropriate type 2 MI code is sequenced as the principal diagnosis, the case would be placed in the MI readmission cohort rather than the heart failure readmission co-hort. Note: Since troponin interpretation is an art, consider an article from the Mayo Clinic on the new high-sensitive troponins available at www.amjmed.com/article/S0002-9343(17)30826-4/fulltext.

– A patient is admitted with a type 2 MI. If the underlying cause of the type 2 MI is clearly docu-mented as contributing to the need for the inpatient admission (e.g., rapid atrial fibrillation), it can be sequenced as the principal diagnosis, removing the case from the MI readmission cohort.

http://hcpro.com

http://tinyurl.com/mmvprzy

http://www.acmaweb.org/Leadership/2017/17LPA_Session5_Kennedy_Notes.pdf

http://www.acmaweb.org/Leadership/2017/17LPA_Session5_Kennedy_Notes.pdf

http://www.bcbsks.com/CustomerService/Providers/Publications/institutional/newsletters/2013/09-10_new-patient-discharge-codes-effective-10-1.htm



http://www.amjmed.com/article/S0002-9343(17)30826-4/fulltext



• Know specific nuances of the exclusion criteria for certain readmission measures CMS has some interesting rules regarding cohort exclusions. These include:

– Pneumonia. While CMS includes sepsis due to pneumonia in the readmission cohort, it excludes se-vere sepsis (sepsis resulting in acute organ dysfunction) or septic shock when it is clearly document-ed by the physician and coded using ICD-10-CM. As such, if any of our pneumonia patients qualify for the SEP-1 core measure, our documentation of severe sepsis or septic shock and what the asso-ciated organ dysfunction is excludes the case from the pneumonia readmission cohort. Pneumonia documented to be due to or associated with HIV disease is also excluded if explicitly documented, provided that ICD-10-CM code B20, HIV disease, is sequenced as the principal diagnosis.

– CABG. CMS excludes certain heart procedures performed with coronary artery bypass grafting (CABG), such as a concomitant valve surgery, from the CABG readmission cohort. One proce-dure that is excluded—but not often documented by thoracic surgeons—is the extensive lysis of mediastinal adhesions impeding access to the left or right ventricle, prolonging operative time, and increasing operative risk that occurs during a second or third heart operation. Coders, how-ever, cannot report this difficult or extensive lysis of adhesions in ICD-10-PCS unless they are explicitly documented by the surgeon as being significant, affecting access to the right or left ven-tricle, not just the heart.

– TIA versus aborted stroke w/tPA. If your facility administers tissue plasminogen activator (tPA) to a patient with acute cerebrovascular ischemia within 4.5 hours of the onset of symptoms, the patient’s symptoms later resolve within 24 hours, and the neuroimaging is subsequently nega-tive, then labeling, coding, and sequencing transient ischemic attack (TIA) instead of “aborted stroke” or “stroke” as a principal diagnosis excludes the patient from the stroke readmission co-hort. Please consider the American Stroke Association’s definition of TIA and stroke in mak-ing this judgment, available at http://stroke.ahajournals.org/content/strokeaha/early/2013/05/07/STR.0b013e318296aeca.full.pdf. You may need to assure your stroke teams that MS-DRGs no longer require documentation of “aborted stroke” in patients with a TIA or acute cerebrovas-cular insufficiency receiving tPA to qualify for the higher weighted DRG. Learn which other diagnoses qualify for the tPA DRG at www.cms.gov/ICD10Manual/version35-fullcode-cms/full-code_cms/P0060.html.

• Get the risk adjustment right In calculating an expected readmission rate, CMS looks at all diagnosis codes during the index admission and those that are collected by physicians and hospitals within the previous 12 months. Unfortunately, for a number of these, if they were not collected prior to the index inpatient ad-mission, they will not count in the risk adjustment. These include certain electrolyte imbalances, hemiparesis, and the like. Consequently, we must encourage and work with all referring providers who care for their patients in their clinics, nursing homes, home visits, and elsewhere to document and code all conditions affecting the expected readmission rate. A rigorous preoperative evalu-ation performed more than three or more days prior to the inpatient order whereby these codes are reported will also help. See the following guidance:

– A PDF table of codes, available at http://tinyurl.com/2015ReadmissionCodes CMS’ official meth-odologies, available at www.qualitynet.org/dcs/ContentServer?c=Page&pagename=QnetPublic%2FPage%2FQnetTier4&cid=1219069856694. Click on the PDF to learn what condition categories


http://hcpro.com

http://stroke.ahajournals.org/content/strokeaha/early/2013/05/07/STR.0b013e318296aeca.full.pdf

http://stroke.ahajournals.org/content/strokeaha/early/2013/05/07/STR.0b013e318296aeca.full.pdf

http://www.cms.gov/ICD10Manual/version35-fullcode-cms/fullcode_cms/P0060.html

http://www.cms.gov/ICD10Manual/version35-fullcode-cms/fullcode_cms/P0060.html

http://tinyurl.com/2015ReadmissionCodes

http://www.qualitynet.org/dcs/ContentServer?c=Page&pagename=QnetPublic%2FPage%2FQnetTier4&cid=1219069856694

http://www.qualitynet.org/dcs/ContentServer?c=Page&pagename=QnetPublic%2FPage%2FQnetTier4&cid=1219069856694



are excluded if only on the initial encounter. Click on the CC to ICD-10 crosswalks to obtain Excel spreadsheets of which codes map to the condition categories affecting risk adjustment.

• Please partner with your HIM, CDI, or case management staff on how best to capture these on an outpatient basis. Provided CMS continues the Merit-based Incentive Payment System (MIPS), this will not only help the hospital, it will also help us with the cost-efficiency portion of our MIPS bonus that will be paid in 2021.

I encourage you to consider the references in this article and your cooperation with your case manage-ment, CDI, and coding teams in addressing this issue. I thank you for your attention.

With kind regards,

James S. Kennedy, MD, CCS, CDIP, CCDS President, CDIMD.com

EDITOR’S NOTEDr. Kennedy is a general internist and certified coder, specializing in clinical effectiveness, medical informatics, and clinical documentation and coding improvement strategies. Contact him at 615-479-7021 or at [email protected]. Advice given is general. Readers should consult professional counsel for specific legal, ethical, clinical, or coding questions. Opinions expressed are that of the author and do not necessarily represent HCPro, ACDIS, or any of its subsidiaries. For any other questions, contact Editor Nicole Votta at [email protected].


The highest-quality, need-to-know information, written by industry veterans

Shop our catalog

http://hcpro.com



http://www.hcmarketplace.com

him briefings - enjoinvolume 33 issue no. 2 february 2018 him briefings clinical documentation...

Documents