| Download This File as a PDF |
|
|
| |
| DRGGROUPERS.COM MANUAL |
| Looking for more information about DRGs, groupers in general, or our
grouper product line in particular?
Download our complete manual / catalog / reference as a PDF:
drgman.pdf. Available soon as a paperback
book. |
| [Top of Page] |
| |
| HOW TO PICK A VERSION |
| US Federal (aka CMS or HCFA) DRG versions are released in October of
every year. Version 3 (the first version we support) was released in
October of 1985. So if you are using calendar years,
1/1/2001 - 9/30/2001 would be covered by version 18 and 10/1/2001
through 9/30/2002 would be covered by version 19. And so on. |
| Year | Jan-Sep | Oct-Dec |
|
Year | Jan-Sep | Oct-Dec |
|
Year | Jan-Sep | Oct-Dec |
| 2009 |
f26 |
f27 |
|
1999 |
f16 |
f17 |
|
1989 |
f6 |
f7 |
| 2008 |
f25 |
f26 |
|
1998 |
f15 |
f16 |
|
1988 |
f5 |
f6 |
| 2007 |
f24 |
f25 |
|
1997 |
f14 |
f15 |
|
1987 |
f4 |
f5 |
| 2006 |
f23 |
f24 |
|
1996 |
f13 |
f14 |
|
1986 |
f3 |
f4 |
| 2005 |
f22 |
f23 |
|
1995 |
f12 |
f13 |
|
1985 |
f2 |
f3 |
| 2004 |
f21 |
f22 |
|
1994 |
f11 |
f12 |
|
1984 |
N/A |
N/A |
| 2003 |
f20 |
f21 |
|
1993 |
f10 |
f11 |
|
1983 |
N/A |
N/A |
| 2002 |
f19 |
f20 |
|
1992 |
f9 |
f10 |
|
1982 |
N/A |
N/A |
| 2001 |
f18 |
f19 |
|
1991 |
f8 |
f9 |
|
1981 |
N/A |
N/A |
| 2000 |
f17 |
f18 |
|
1990 |
f7 |
f8 |
|
1980 |
N/A |
N/A |
Notes
- We do not support version 2
- Current version is f27
|
|
| [Top of Page] |
| |
| WHAT ARE DRGS? |
| DRG stands for "Diagnosis Related Group."
DRGs are small integers ranging from 0 to about 500. These integers represent inpatient
classifications on the basis of diagnosis, procedure, age, gender and discharge disposition.
These groups were constructed to control Length-of-stay, which in turn correlates to resource
consumption and severity of illness. |
| There are two special DRG values: 0 (which means "not grouped")
and 470 (which means "ungroupable"). The rest of the DRG values have descriptions, weights,
LOS outlier trim points and mean LOS all of which depend on the DRG version. |
| Wikipedia has a pretty good entry on
DRGs. |
| [Top of Page] |
| |
| HCFA VERSUS CMS |
| In 2001,
the United States federal government's Health Care Finance Administration
("HCFA")
became "the Centers for Medicare & Medicaid Services" or "CMS". |
| [Top of Page] |
| |
| WHERE DID DRGS COME FROM? |
| The original
DRGs were invented at Yale University's Health Systems Management Group in the
late 1970s. The principal researchers were Bob Thompson, a nursing guru, and
Bob Fedder, an Operations Research kind of guy. Ron Mills, co-founder of the
parent of DRGGroupers.com, was
the technical lead and he was the one who created the biostatistical analysis
package, AUTOGRP, which made the underlying research possible in real-time. |
| DRGs were adopted by
the United States federal government's Health Care Finance Administration
(HCFA) and first released in 1982 as version 2
(version 1 was the unreleased version which HCFA evaluated).
Every year, on October 1st, HCFA (now CMS) releases a new CMS DRG version. |
| Through Ron Mills, DRGGroupers.com has been involved with DRGs from their inception.
Ron wrote the first grouper (a program which assigns a DRG to an inpatient encounter)
in 1978 which is the model for the grouper CMS (formerly HCFA) still uses today. |
| [Top of Page] |
| |
| WHAT ARE DRGS GOOD FOR? |
| DRGs are good for providing a context in which to analyze hospital stays. DRGs were designed
to allow hospitals to operate on a more industrial basis, with resource allocation and
cost-center analysis, all of which were very hip in the late 1970s when DRGs were created.
In a nutshell, DRGs predict likely resource consumption for any given hospital stay, allowing
one to determine if the given hospital stay was too short, too long or just right. |
| [Top of Page] |
| |
| WHAT ABOUT REIMBURSEMENT? |
| Inherently, DRGs have nothing to do with reimbursement. But by historical accident,
DRGs were chosen by Medicare as the basis
of the Prospect Payment System for hospitals. Since DRGs hit the scene as part of a reimbursement
scheme, DRGs became linked with reimbursement in many people's minds. |
| [Top of Page] |
| |
| WHAT IS A PRICER? |
| Since DRGs measure resource consumption in the form of a normalized weight, using DRGs for
reimbursement not only makes sense, it is easy: you multiply the DRG-specific weight by the
facility-specific factor and voila! you have a reimbursement amount for a given inpatient stay.
However, this addition step is called "pricing" and is not part of the grouper per se; it is
a separate process which is not part of grouping. Software which makes this calculation is called
a "pricer." For convenience, most pricer providers bundle the DRG grouper in with their software,
which had confused grouping and pricing in many people's minds. |
| We are hardly experts on buying pricing software, but if you are looking to buy it and are
stuck, check out Health Information Systems (a division of 3MTM). They seem to have lots of pricers
out there in the world, so someone is buying them. |
| [Top of Page] |
| |
| WHAT ABOUT CODING DIAGNOSES AND PROCEDURES? |
| The official grouper only accepts
ICD9cm codes (International
Committee on Diseases, version 9, Clinical Modifications) for both diagnoses
and procedures. However, the
American Medical Association has defined
an alternative scheme for coding procedures, which they call
CPT (Current Procedural
Terminology). Many providers have chosen to code even in-house procedures using
CPT. But if you want to group with CPT codes as input, then you have to convert
them to ICD9cm codes first. This conversion is not a simple one-to-one mapping. Many
vendors sell CPT-to-ICD9cm "crosswalks," but DRGGroupers.com is not one of them. |
| [Top of Page] |
| |
| WHAT IS A GROUPER? |
| A DRG Grouper is a computer program or module which takes those 5 clinical and
demographic data as input and gives a corresponding Diagnosis Related Group
as output. The diagnoses and procedures are encoded as
ICD9cm codes (International
Committee on Diseases, version 9, Clinical Modifications).
The age is a small integer from 0 to 129.
The gender is encoded
as 1 for male, 2 for female and 3 for unknown (don't ask). The discharge disposition,
also known as "discharge status,"
is usually encoded either using UHDDS or UB82 (both medical billing standards). |
| For a PDF from CHIMA common discharge status codes,
follow this link or go straight to the source at
http://www.chima.org/DischDispCodesReference2005Update.pdf
|
| The standard CMS (formerly HCFA) grouper, ours included, will accept up to 10 diagnoses, which are
presumed to be
in order of significance, from the Primary diagnoses (number 1) on down the line.
Likewise, up to 15 procedures are accepted, but their significance is determined by
the grouping process, so their order is not important. |
|
| The relevance of any diagnosis or procedure code is determined by its mask,
which is a bitstring of conditions. The masks guide the grouper in its use of any
given code; for instance, the masks say whether or not a code is gender-specific,
or if it is allowed as a primary diagnosis. In addition to the information
encoded in the mask,
the grouper applies logic to actually classify any given inpatient stay
into a single DRG. |
| [Top of Page] |
| |
| WHAT ARE DRG PROPERTIES? |
| Any inpatient stay can be classified into a CMS (formerly HCFA) Diagnosis Related Group. Any CMS DRG
for a given version
has certain properties, determined empirically by CMS from the MedPar database.
Those properties are: |
- A DRG description (70 characters wide, version-dependant)
- An MDC (see below for details)
- A Geometric Mean Length of Stay (GMLOS)
- A Weight (a normalized prediction of resource consumption)
- A Category: either "Surgical" or "Medical"
- A low "trim point" (the LOS below which lie the low outliers)
- A high "trim point" (the LOS above which lie the high outliers)
|
| The DRGGroupers.com grouper returns all these and more: a bit string for each of the Diagnosis
Codes and Procedure Codes so that the caller can determine which codes were actually
significant to the grouping. |
| [Top of Page] |
| |
| WHAT ARE MDCS? |
| MDC stands for "major diagnostic category" and is a sort of pre-DRG classification of
inpatient stays. |
| MDC | Description |
|---|
|
1 |
Diseases & Disorders of the Nervous System
|
|
2 |
Diseases & Disorders of the Eye
|
|
3 |
Diseases & Disorders of the Ear, Nose, Mouth & Throat
|
|
4 |
Diseases & Disorders of the Respiratory System
|
|
5 |
Diseases & Disorders of the Circulatory System
|
|
6 |
Diseases & Disorders of the Digestive System
|
|
7 |
Diseases & Disorders of the Hepatobiliary System & Pancreas
|
|
8 |
Diseases & Disorders of the Musculoskeletal System & Conn Tissue
|
|
9 |
Diseases & Disorders of the Skin, Subcutaneous Tissue & Breast
|
|
10 |
Endocrine, Nutritional & Metabolic Diseases & Disorders
|
|
11 |
Diseases & Disorders of the Kidney & Urinary Tract
|
|
12 |
Diseases & Disorders of the Male Reproductive System
|
|
13 |
Diseases & Disorders of the Female Reproductive System
|
|
14 |
Pregnancy, Childbirth & the Puerperium
|
|
15 |
Newborns & Other Neonates with Condtn Orig In Perinatal Period
|
|
16 |
Diseases & Disorders of Blood, Blood Forming Organs, Immunolog Disord
|
|
17 |
Myeloproliferative Diseases & Disorders, Poorly Differentiated Neoplasm
|
|
18 |
Infectious & Parasitic Diseases, Systemic or Unspecified Sites
|
|
19 |
Mental Diseases & Disorders
|
|
20 |
Alcohol/drug Use & Alcohol/drug Induced Organic Mental Disorders
|
|
21 |
Injuries, Poisonings & Toxic Effects of Drugs
|
|
22 |
Burns
|
|
23 |
Factors Influencing Hlth Stat & Othr Contacts with Hlth Servcs
|
|
24 |
Multiple Significant Trauma
|
|
25 |
Human Immunodeficiency Virus Infections
|
|
| [Top of Page] |
| |
| IS THERE AN OFFICIAL GROUPER? |
| In the United States of America, the "official" is the one defined by CMS (formerly HCFA).
Strictly speaking, the CMS grouping algorithm is public, and anyone can implement it in software.
(There are books published so that one could even do without the software and assign DRGs by
hand.)
However, CMS has blessed Health Information Systems (a division of 3MTM) as the distributor of the
reference grouper, which is written
in IBM 360 Mainframe assembler. If you have an IBM 360-compatible computer, you can buy that
grouper through 3MTM-HIS and run that. |
| (Historical Note: Ron Mills also founded a company which became Health
Systems International, which became HIS when it was bought by 3MTM.) |
| [Top of Page] |
| |
| WHAT ARE RDRGS®? |
| Since DRGGroupers.com is constantly asked about RDRGs®, we asked the nice folks at HSC to give
us a blurb to put on our website to answer this question. Here is their reply: |
|
The RDRG severity-of-illness software is a product of Health Systems Consultants, Inc. in
New Haven, Connecticut. The software groups inpatient hospital discharge data into DRGs and into
severity classes within DRGs. The DRGs produced are identical to those of the public domain DRG grouper
from the Health Care Financing Administration (HCFA--now CMS). The software assigns patients to 511 DRGs and to
1198 Refinement Group (RGN) numbers and is updated each year to conform to the CMS DRGs. Since the software
system can predict hospital resource use, it can be used to improve hospital casemix analysis, analyze
hospital performance, evaluate physician performance, measure quality, develop budgets, and to reimburse hospitals.
|
|
The RDRG severity-of-illness software was developed from a Yale University study funded by CMS (formerly HCFA)
entitled, "DRG Refinement with Diagnostic Specific Comorbidities and Complications: A Synthesis
of Current Approaches to Patient Classification." The study, completed in 1989, was designed to adjust
the DRG system for the severity of a patient's illness. For information about the RDRG software,
please contact Karen Schneider at
karen.Schneider@healthsyst.com
or call Health Systems Consultants at (203) 785-0650.
|
| [Top of Page] |
| |
| WHO BUYS THE OUR GROUPERS? |
| Our market niche is batch grouping under UNIX or under MS-DOS and MS-Windows. Our clients are mostly
health care consulting
firms or the IS divisions of HMOs. Since we are not primarily a software house, we are not set up to
provide any support beyond the basic installation and trouble-shooting. If the idea of a UNIX-style filter
is unknown to you or is scarey to you, you don't want to buy a grouper from us. |
| On the other hand, if you want a grouper that returns all the DRG-related information about an inpatient
stay and which provides a list of significant codes, then you might want to buy a grouper from us. Visit
our Grouper products and prices page for more information. |
| [Top of Page] |
| |
| WHO DEFINES DRGS? |
| While we usually mean "US Federal DRGs" when we say "DRGs,"
there are many different governments which have defined their own version of DRGs.
New York state defined their own. New Jersey defined their own for a while. France has
their own, as does Portugal. Australia recently joined the club with their own
version. |
| [Top of Page] |
| |
| WHY SO MANY KINDS OF DRGS? |
| The creators of the
CMS (formerly HCFA) DRGs were constrained by the number of data elements CMS felt that they could reasonably
expect any given hospital in the country to collect.
Furthermore, their baseline population is all Medicare patients, which skews the results
somewhat. As a result, the CMS DRGs are unambitious with respect to severity of illness
and resource consumption and not appropriate to all hospital populations. |
| Many groups have tried to extend the basic DRG concept to fix these flaws.
3MTM/HIS sells AP-DRGs ("All Payor" DRGs). Yale University's School of Medicine came up with
RDRGs ("Refined" DRGs). CMS itself is working on SDRGs ("Severity-adjusted" DRGs). |
| [Top of Page] |
| |
