STATE OF NORTH CAROLINA
IN THE OFFICE
OF ADMINISTRATIVE HEARINGS
COUNTY OF GUILFORD
05 DHR 0719
______________________________________________________________________________
FELICIA BOYKIN, RHIA,
MOSES CONE HOSPITAL,
Petitioner,
)
)
)
)
v.
)
DECISION
)
NORTH CAROLINA DEPARTMENT of
)
HEALTH and HUMAN SERVICES,
)
DIVISION OF MEDICAL ASSISTANCE, )
Respondent.
)
______________________________________________________________________________
THIS MATTER came on for hearing before the undersigned Administrative Law Judge,
Augustus B. Elkins II, on September 1, 2005 in High Point, North Carolina.
APPEARANCES
For Petitioner:
Melanie M. Hamilton
Ott Cone & Redpath, P.A.
1501 Highwoods Boulevard, Suite 101
Greensboro, North Carolina 27410
For Respondent: Brenda Eaddy
Assistant Attorney General
N.C. Department of Justice
9001 Mail Service Center
Raleigh, North Carolina 27699-9001
ISSUE
Whether Petitioner improperly coded the medical stay of a medical recipient and whether
Respondent acted correctly when it issued its letter notifying Petitioner of its proposed
recoupment based on its amended diagnosis, resulting in a lesser allowed payment.
EXHIBITS
The record on this case was sealed at the request of the parties to protect the patient’s
confidentiality
Petitioner’s exhibit 1 which is the lengthy patient medical record was admitted.
-1-
Respondent’s exhibits 1 through 14 were admitted.
BASED UPON careful consideration of the sworn testimony of the witnesses presented
at the hearing, the documents and exhibits received and admitted into evidence, and the entire
record in this proceeding, the Undersigned makes the following findings of fact. In making the
findings of fact, the Undersigned has weighed all the evidence and has assessed the credibility of
the witnesses by taking into account the appropriate factors for judgment of credibility, including
but not limited to the demeanor of the witnesses, any interests, bias, or prejudice a witness may
have, the opportunity of the witness to see, hear, know or remember the facts or occurrences
about which the witnesses testified, whether the testimony of the witness is reasonable, whether
the testimony is consistent with all other believable evidence in the case, and the qualifications of
the witness as an expert.
FINDINGS OF FACT
1. Petitioner is a hospital and provider of medical services to Medicaid recipients. The
Respondent (DMA) is the state agency responsible for administering and managing the State
Medicaid Plan and Program.
2. This matter is before the Undersigned due to a recoupment action. Respondent has requested
Petitioner re-pay $10,802.68 back to Respondent for improper coding of the hospital stay of a
Medicaid recipient.
3. On September 29, 2004, the Medicaid recipient in this action was transported to Petitioner’s
hospital facility via ambulance. This Medicaid recipient remained in the hospital until
October 4, 2004. The discharge summary listed the final diagnosis according to the primary
care physician. The final discharge diagnosis shows “ventilator-dependent respiratory
failure” as the first listed diagnosis. Respondent initially made payment to Petitioner in the
amount of $14,431.23 for the care and treatment of the patient.
4. The claim as submitted by Petitioner with the assigned DRG of 475 (Respiratory System
Diagnosis with Ventilator Support) was based on the following diagnosis and procedure
codes: PDX1 - 51881 Acute Respiratory Failure; DX1 – 481 Pneumococcal Pneumonia;
DX2 – 496 Chronic Airway Obstruction NEC; DX3 – 07030 Hepatitis B Acute without
Coma; DX4 – 4169 Chronic Pulmonary Heart Disease NOS; DX5 – 4019 Hypertension
NOC; DX6 – 41400 Coronary Atherosclerosis Unspecified Vessel Native/Graft; DX7 –
V4581 Aortocoronary Bypass; DX8 – 412 Old Myocardial Infarct; and PX1 – 9671
Continuous Mechanical Ventilation for less than 96 hours.
5. In a post payment review, Respondent coded this hospital stay showing poisoning by
benzodiazepine as the primary diagnosis. By letter dated January 18, 2005, Respondent, by
and through its agent Medical Review of North Carolina (MRNC), issued a request for
recoupment contending Petitioner had been overpaid in the amount of $10,802.68 due to an
inappropriate principal diagnosis code on the claim submitted for payment. Respondent’s
letter dated January 18, 2005, notified Petitioner of its proposed recoupment based on
-2-
amended diagnosis coding of PDX 9694 Poisoning- Benzodiazepine Tranquilizers and DX 1
96509 Poisoning - Opiates NEC, with all subsequent diagnoses following those listed by
Petitioner. Under the ‘amended coding’ made by Respondent, the resultant DRG changed to
449 – Poisoning and Toxic Effects of Drugs Age Greater Than 17 With Comorbid
Conditions, and a lesser allowed payment
6. Shawnee Gatling is an expert in health information administration with expertise in coding.
Ms. Gatling is an employee of Medical Review of North Carolina (MRNC). One of
MRNC’s responsibilities is to perform post-payment reviews of the payment records of
Medicaid providers in order to determine if the hospital stays of Medicaid recipients have
been coded correctly for payment. Ms. Gatling testified as to her review of the coding in this
matter. Coding Clinic issue 2002, 4th quarter states that if a diagnosis at the time of discharge
is probable or not ruled out, the condition should be coded as if it existed or was established.
Coding Clinic issue 1991, 2nd quarter sets forth the sequencing of respiratory failure when a
non-respiratory condition presents also. Sub paragraph 3 states that when a patient is
admitted with respiratory failure due to or associated with an acute non respiratory condition,
the acute condition is sequenced as the principal diagnosis. In this case, Ms. Gatling found
that the overdose should be coded first since she did not find that it had been specifically
ruled out, and should be coded as a poisoning.
7. The stated reason for the recipient’s admission to the hospital as noted in the Petitioner’s
Health History/Assessment was: respiratory failure; “took too many pills” - accident. In
addition, the diagnosis listed in the physician’s orders for that day, included the words
“respiratory failure - overdose”. Under ‘Assessment and Plan’ in the Addendum to the H &
P dated September 30, 2004, the doctor noted that one of the causes of this patient’s
respiratory depression was drug overdose, and one of the likely causes of his respiratory
pressure was drug induced respiratory depression. However, the final discharge summary
did not list “overdose” or “poisoning” as a diagnosis.
8. Respondent contends that two additional diagnoses (9694 Poisoning - Benzodiazepine
Tranquilizers and 96509 Poisoning - OPICS NEC), should be added and listed as the
principal and first diagnosis for this patient.
9. Carolyn Bennett is an in-patient coder for Moses Cone Hospital where she has been for over
18 years. Testimony from Ms. Bennett is consistent with Dr. Samuel Cykert, the attending
physician. She testified that she interpreted the doctor’s decision not to list overdose or
poisoning as the doctor having ruled out poisoning or overdose as the cause of the patient’s
admission to the hospital. She further stated it was the finding from the discharge summary
that it had been ruled out based upon a discussion in the discharge summary that the patient
adamantly denied taking an overdose, that only two pills were missing from his prescription
bottle, and that overdose had not been confirmed as a diagnosis. Furthermore, as of the time
of the patient’s discharge, the whole record did not indicate that poisoning was “probable,
suspected, likely, questionable, possible or still to be ruled out.”
10. Ms. Bennett testified that one of the primary reasons for coding an uncertain diagnosis is in
cases where no diagnosis could otherwise be coded. She testified that there are
-3-
circumstances (either because a patient leaves against medical advice, the patient dies or for
other reasons) where no confirmed diagnosis can be made at the time of discharge. In those
circumstances, uncertain diagnoses must be coded in order to obtain any payment. That is
very different from this case where there were multiple identified diagnoses at the time of the
patient’s discharge.
11. The oral testimony of Carolyn Bennett and Shawnee Gatling establishes that individual
coders will code in different ways. Both women testified that they code their claims
differently from other coders. Some individuals may code from the beginning of the
patient’s treatment to the end of their treatment, while others may look primarily to the
patient’s discharge summary for diagnoses, with further support based on the entirety of the
medical record. Ultimately, the primary purpose of coding is to list the reasons the patient
was in the hospital.
12. Dr. Samuel Cykert testified for the Petitioner. He was the attending physician of the patient.
He is Board certified and has been at Moses Cone for approximately 14 years. It is
undisputed that the treating physicians did not list poisoning as a diagnosis for this patient
upon his discharge. It is further undisputed that this patient was never diagnosed with having
overdosed or been poisoned. When the patient was first admitted to the hospital a question
arose as to whether he may have overdosed, partly because the patient was unable to speak
for himself. By the time of his discharge, there was insufficient evidence to justify a
diagnosis of overdosing or poisoning. To the contrary, Dr. Cykert testified that the treatment
and testing provided to the patient proved that it wasn’t just that he couldn’t breathe because
he was unconscious, but the testing also proved that he had intrinsic lung disease causing his
respiratory failure.
13. Dr. Cykert testified that the diagnosis listed first on the discharge summary is the patient’s
principal diagnosis and the true reason the patient was hospitalized in the first place. In this
case, the principal diagnosis assigned for the patient was ventilator dependent respiratory
failure. When questioned whether he believed poisoning to be an appropriate diagnosis, Dr.
Cykert stated that while the patient had benzodiazepines in his urine, no level was indicated,
and the patient was taking benzodiazepines under his regular prescription.
14. After the patient had been taken off of the ventilator, he denied taking any extra medicine
that would indicate an overdose. Further, his wife claimed that she knew how many pills
were in his prescription bottle and that the patient had not taken any extra medication.
15. The patient’s physicians consciously “dropped intoxication as a diagnosis” in the discharge
summary because of the history from both the person who presented unconscious and from
the wife who lived with him. Dr. Cykert stated that even if poisoning were a diagnosis, it
should not have been the principal diagnosis because even if it had contributed to his
admission, it was not the only thing and not the primary reason that caused his respiratory
failure.
-4-
BASED UPON the foregoing findings of fact and upon the preponderance or greater
weight of the evidence in the whole record, the Undersigned makes the following:
CONCLUSIONS OF LAW
1. The N.C. Office of Administrative Hearings has jurisdiction over the parties and subject
matter of this contested case pursuant to N.C.G.S. 150B-23, et. seq., and there is no question
as to misjoinder or nonjoinder. The parties received proper notice of the hearing in the
matter. To the extent that the findings of fact contain conclusions of law, or that the
conclusions of law are findings of fact, they should be so considered without regard to the
given labels.
2. The Respondent has not raised any issue regarding the appropriateness of the diagnosis and
other codes that were submitted by the Petitioner. Respondent contends that two additional
diagnoses (9694 Poisoning - Benzodiazepine Tranquilizers and 96509 Poisoning - OPICS
NEC), should be added and listed as the principal and first diagnosis for this patient.
Because it is undisputed that if one of these “additional” diagnoses had not been listed as the
principal diagnosis, the payment made to Petitioner would not have changed, the ultimate
issue is not whether these codes should have been coded on this patient’s claim to Medicaid,
but whether one of them should have been listed as the principal diagnosis.
3. The language of the guideline requiring uncertain diagnoses to be coded as though they
existed does not apply to this patient in that, at the time of his discharge it was not “probable,
suspected, likely, questionable, possible, or still to be ruled out.” Moreover, even if it were
coded as an uncertain diagnosis under a guideline, there is no guideline requiring that a
“possible” or “uncertain diagnosis” be coded higher in the hierarchy than a confirmed
diagnosis.
4. The evidence does not support that coding guidelines require that if poisoning had been
coded, because it had not been “ruled out,” that it would carry the same weight as the
patient’s actual diagnoses reached by the physician and medically trained personnel.
5. The purpose of coding a claim for Medicaid is to identify the true reason a patient was
admitted because it impacts the type of treatment provided to the patient, and further
determines the payment Medicaid will make to the hospital. Coding Clinic, Fourth Quarter
2002, Principal Diagnosis For Inpatient, Short Term, Acute Care 2002, specifically states that
the principal diagnosis is defined in the Uniform Hospital Discharge Data Set (UHDDS) as
“that established after study to be chiefly responsible for occasioning the admission of the
patient to the hospital for care.” Subsection A states that “Codes for symptoms, signs, and
ill-defined conditions from Chapter 16 are not to be used as the principal diagnosis when a
related definitive diagnosis has been established.”
6. Respondent relies upon Subsection H regarding Uncertain Diagnoses, which states: “if the
diagnosis documented at the time of discharge is qualified as “probable,” “suspected,”
“likely,” “questionable,” “possible,” or “still to be ruled out,” code the condition as if it
existed or was established. In this case, a “diagnosis” of overdose or poisoning was not
-5-
qualified by the physician in the record as probable, suspected, likely, questionable, possible,
or still to be ruled out at the time of the patient’s discharge. Rather, the treating physician
consciously chose not to list poisoning or overdose in the discharge summary. Instead, the
treating physicians identified the diagnoses which in their medical opinion were chiefly
responsible for occasioning the patient’s admission to the hospital for care.
7. Section 3 of the February, 2004 Coding Guidelines and the 2005 ICD-9-CM relate to the
reporting of Additional Diagnoses For Inpatient, Short Term, Acute Care And Long Term
Care Hospital Records. Under the section regarding General Rules for Other (Additional)
Diagnoses it states that for reporting purposes the definition for “other diagnoses” is
interpreted as additional conditions that affect patient care in terms of requiring clinical
evaluation, or therapeutic treatment; or diagnostic procedure; or extended length of hospital
stay, or increased nursing care and/or monitoring. The evidence in this case established that
no additional resources were expended related to any “suspected” poisoning.
8. The primary coding clinic relied on by Respondent to state that poisoning should have been
listed as the principal diagnosis for this patient is the Second Quarter 1991 “Respiratory
Failure With Non-Respiratory Conditions - Guidelines.” However, the “question” and all of
the examples identified in that coding clinic presuppose that a patient has actually been
diagnosed as having a specific condition. In this case, the patient was never diagnosed as
having overdosed or been poisoned, nor was any finding made that the benzodiazepines in
his system led in anyway to his respiratory condition which brought him to the hospital.
Therefore this guideline does not mandate that poisoning be listed as the principal diagnosis
for this patient.
9. Respondent also relied upon the Coding Guidelines identified in the ICD-9-CM Expert
Manual for 2005 set forth on page 9, paragraph 2 regarding poisoning. However, the weight
of the evidence shows that each of the situations identified therein do not apply to this
patient. Subsection A applies when an error is made in a drug prescription or in the
administration of the drug, but there is no evidence that an error was made in either the
prescription or administration of benzodiazepines for this patient. Subsection B applies when
an overdose of a drug was intentionally taken or administered, of which there is no evidence
in this case. Subsection C relates to a non-prescribed drug or medicinal agent, yet it is clear
that this patient was prescribed benzodiazepine. The last Section D “states when coding a
poisoning or reaction to the improper use of a medication (e.g., wrong dose, wrong
substance, wrong route of administration) the poisoning code is sequenced first, followed by
a code for the manifestation.” In this case, there was no definitive basis to code a poisoning
because none of the matters set forth in A through C were present, nor was it identified to
have been a reaction to an improper use of a medication.
10. Respondent replied upon Coding Clinic 1993 First Quarter for Overdose which had a
question asking “how do you code the patient who is diagnosed as overdosing on crack and
found to be in respiratory failure, then placed on ventilation? Wouldn’t the poisoning code
be the principal diagnosis?” The example set forth in the question and answer both
presupposes that the patient has been diagnosed as overdosing. Such is not the case in this
-6-
matter and therefore this coding clinic does not serve to mandate that the principal diagnosis
for this patient be poisoning.
11. The evidence does not support finding that an unconfirmed or uncertain diagnosis should be
put on equal or higher position with confirmed diagnoses. The evidence of this record does
not support a mandate that a suspected diagnosis initially set forth when the patient was
unconscious and first admitted, be coded as the primary reason the patient was admitted to
the hospital, when, after full and thorough testing and review of the patient, the true diagnosis
and reason for admission has been revealed.
12. The preponderance of the evidence in this case supports and holds that the reason the patient
was admitted to the hospital was respiratory failure. If and to the extent poisoning was
related at all, it did not occasion the patient’s admission and therefore should not be listed as
the principal diagnosis for this patient.
BASED UPON the foregoing Findings of Fact and Conclusions of Law, the Undersigned
makes the following:
DECISION
There is sufficient evidence to properly and lawfully support that the diagnoses assigned
by Petitioner were proper and appropriate. Applicable coding guidelines do not require that
Petitioner have coded poisoning as a diagnosis for this patient when in fact the preponderance of
the evidence shows that the reason the patient was admitted to the hospital was respiratory
failure. If and to the extent it was appropriate to code poisoning as a diagnosis for this patient
under the theory that poisoning had not been “ruled out” by the patient’s physician at the time of
his discharge, such codes should not have been listed as the patient’s principal diagnosis since
the physician had documented actual diagnoses for the patient as having caused his admission to
the hospital. The Respondent’s Notice of Recoupment was in error, and Petitioner is entitled to
an additional $10,802.68 in reimbursement for the services rendered to this patient.
NOTICE
The agency making the final decision in this contested case shall adopt the Decision of
the Administrative Law Judge unless the agency demonstrates that the Decision of the
Administrative Law Judge is clearly contrary to the preponderance of the admissible evidence in
the official record. The agency is required to give each party an opportunity to file exceptions to
this Decision issued by the Undersigned, and to present written arguments to those in the agency
who will make the final decision. N. C. Gen. Stat. § 150B-36(a).
In accordance with N.C. Gen. Stat. § 150B-36, the agency shall adopt each finding of fact
contained in the Administrative Law Judge’s decision unless the finding is clearly contrary to the
preponderance of the admissible evidence, giving due regard to the opportunity of the
Administrative Law Judge to evaluate the credibility of witnesses. For each finding of fact not
adopted by the agency, the agency shall set forth separately and in detail the reasons for not
adopting the finding of fact and the evidence in the record relied upon by the agency. Every
-7-
finding of fact not specifically rejected as required by Chapter 150B shall be deemed accepted
for purposes of judicial review. For each new finding of fact made by the agency that is not
contained in the Administrative Law Judge’s decision, the agency shall set forth separately and
in detail the evidence in the record relied upon by the agency establishing that the new finding of
fact is supported by a preponderance of the evidence in the official record.
The agency that will make the final decision in this case is the North Carolina
Department of Health and Human Services. The agency is required by N.C.G.S. 150B-36(b) to
serve a copy of the final decision on all parties and to furnish a copy to the parties’ attorneys of
record and to the Office of Administrative Hearings.
IT IS SO ORDERED.
This the 20th day of December, 2005.
___________________________
Augustus B. Elkins II
Administrative Law Judge
-8-