For inpatient clinical documentation to have integrity, it must be accurate, timely, and
completely reflect the patient’s full clinical presentation and scope of services provided for
Complete and accurate documentation helps to facilitate patient care and improved outcomes, and
helps to justify medical necessity for the patient’s stay and services provided.
After inpatient documentation is completed, post-discharge, it is translated into a final coded
abstract. Inpatient cases are classified by diagnostic severity, using Diagnosis Related Groups
(DRGs). Appropriate inpatient coding and DRG classification depends upon an accurately
documented and fully specified primary diagnosis (the reason for admission, after study), and
documentation of all significant comorbidities.
Abstracted coded data is used for many purposes, including clinical research on patient
populations, public health reporting, reputational rankings and quality models. Coded data based
on clinical documentation also directly impacts operational metrics, including Mortality Index
and Length of Stay Variance.
For accurate capture of patient severity, secondary diagnoses and their significance/plan of care
should be included in documentation, if pertinent to the patient’s admission. Significant
diagnoses are those that are measured, monitored, assessed, evaluated or treated, or cause the
patient’s stay to be extended.
Unlike coding for outpatient services, diagnoses included in inpatient documentation that are
“possible”, “probable”, “likely”, “suspected”, “questionable” or “still to be ruled out” at the
time of discharge are included on the final inpatient coded abstract. These are coded for
inpatient cases because they help support medical necessity of the decision for admission, and
are important to properly classify patients according to diagnostic severity.
NYP’s CDI Departments are charged with the mission of helping to capture the most full and
accurate coded abstract possible for each patient stay. Traditionally, this has been approached
by conducting manual chart reviews, sending providers queries to clarify medical record
documentation, and conducting targeted inpatient documentation education sessions for individual
providers and small groups.
New Natural Language Processing (NLP) technology is now available to assist with clinical
documentation integrity as Physicians and APPs document in NYP’s Epic electronic medical record.
Appropriate, pre-defined CDI concepts are surfaced real-time, while documentation is being
created. This provides timely, consistent feedback to a broader clinical audience than was
previously possible with CDI’s manual chart review and query processes.
CDI Tips on Documenting:
Hierarchical condition categories (HCCs) are used by Medicare for risk adjustment.
By Richard D. Pinson, MD, FACP
Developed in 2000, HCCs are part of a risk-adjustment model allowing Medicare to project the
expected future annual cost of care. They're used for calculating payments to Medicare Advantage
plans, accountable care organizations (ACOs), and certain Affordable Care Act (ACA) plans. Many
chronic conditions are included.
Risk adjustment allows Medicare to “level the playing field” so plans that cover patients with
more severe, complex, and costly conditions receive a larger capitated payment than plans with
less costly patients.
HCCs group together ICD-10 codes for related diagnoses with similar clinical complexity and
expected annual costs of care. Each HCC is assigned a relative weight proportional to the
relative costs associated with its constituent diagnoses.
Higher-cost HCCs have higher relative weights. HCC relative weights are therefore similar to
diagnosis-related group weights and to relative value units for CPT codes.
gives a few CMS-HCC examples with relative weights and the number of constituent diagnoses.
Medicare calculates a Risk Adjustment Factor (or RAF, pronounced “raf” as in “raft” without the
“t”) for each patient by combining relative weights for certain of the patient's demographic
factors with the weights of all HCCs covering diagnoses submitted on Medicare claims for that
patient from certain sites of service during the calendar year. The individual patient's RAF
scores are then averaged and this average RAF is multiplied by the base payment rate established
by Medicare for the organization.
An HCC will not be included if one of its constituent diagnoses is not included. Each HCC is
included only once in the RAF calculation. Once a diagnosis from an HCC has been submitted,
other diagnoses in the same HCC have no impact.
RAF calculations are derived from claims submitted for physician offices and hospital inpatient
and outpatient departments.
Today, HCCs are also used for risk adjustment of many quality and pay-for-performance measures
for clinicians and hospitals, including the Merit-based Incentive Payment System (MIPS), the
Hospital Value-based Purchasing Program (VBP), the Hospital Readmissions Reduction Program
(HRRP), and the Hospital-Acquired Condition Reduction Program (HACRP). Hence, it is important to
capture all diagnoses comprising HCCs from the RAF sites of service (including hospitals) and
ensure assignment of the correct RAF.
Dr. Pinson is a certified coding specialist, author, educator, and cofounder of Pinson and Tang, LLC, and is based in Chattanooga,
Tenn. This content is adapted with permission from Pinson and Tang, LLC. The views expressed in
this column are those of the author and not intended to replace authoritative sources for
documentation and coding.
Information referenced from acphospitalist.org
Open Notes refers to the automatic release of visit notes to patients and their proxies via the
NYP Connect Patient Portal, unless a clinician blocks the release due to allowable exceptions to
prevent physical harm or protect patient privacy.
The goal of Open Notes is to improve information transparency, empower patients to engage in
their care, and improve communication between patients and providers.