A Cheaper Hospital In Five Days

Get a Cheaper (More Profitable)

Hospital in Five Days A Special Report by Jay Arthur

It should come as no surprise that a faster, better hospital will be cheaper to operate and more

profitable. When you’re not dealing with all of the delays in the ED-Admission-Discharge

process, fewer patients will be boarded in the ED, reducing diversion and LWOBS. More

patients can be seen more quickly, increasing revenue. When you’re not dealing with the extra

costs of preventable falls, infections and medication errors, it will make the hospital more cost

effective and profitable.

Faster + Better = Cheaper and More Profitable!

And there are other opportunities. Most hospitals have too many problems with rejected,

appealed and denied claims costing millions! Lean Six Sigma can help reduce billing problems

among other operational problems. And the process is simple.

To reduce rejected, appealed and denied claims, use Six Sigma tools to focus the improvement

effort.

1. Analyze Claims using control charts and pareto charts

• Rejected

• Appealed

• Denied

© 2010 KnowWare International Inc. 888-468-1537 1 [email protected]

2. Analyze the Root Causes using the “Dirty 30 process”

3. Implement Countermeasures

4. Track Results

Reducing Denied Claims In Five Days Denied claims mean no money for services rendered because the billing process failed in some

way. Non-payment drives up the cost of healthcare and pushes many hospitals toward

bankruptcy. In this case study, monthly denials were over $1 million (XmR chart).

Charges Coded as Denials

$2,552,122UCL

CL $1,071,509

LCL $(409,103)

$(1,000,000)

$(500,000)

$-

$500,000

$1,000,000

$1,500,000

$2,000,000

$2,500,000

$3,000,000

10/02 11/02 12/02 01/03 02/03 03/03 04/03 05/03 06/03 07/03 08/03 09/03

2002-2003

Cha

rges

Denials (E)UCL+2 Sigma+1 SigmaAverage-1 Sigma-2 SigmaLCL


Using Pareto Charts of Denials Using Excel PivotTables and the QI Macros, it was easy to narrow the focus to a few key areas

for improvement: Timely Filing (61%) and one insurer (67% of Timely Filling denials):

Denial-No Appeal Charges

$150$109,813$336,270$750,766

$1,295,032

$2,516,508

$7,849,569

100%99%97%

91%

81%

61%

$-

$1,607,264

$3,214,527

$4,821,791

$6,429,054

$8,036,318

$9,643,582

$11,250,845

$12,858,109

Timely Filing MedicalNecessity

No Auth Partial Auth Invalid Auth ET ES

Memo Code

Am

ount

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%n=12858108.93

Denials for Timely Filing by Insurer

111447111415182424274581

501

100%100%100%99%99%98%96%94%93%90%87%

84%81%

75%

64%

0

97.25

194.5

291.75

389

486.25

583.5

680.75

778

Ins 1 Ins 2 Ins 3 Ins 4 Ins 5 Ins 6 Ins 7 Ins 8 Ins 9 Ins 10 Ins 11 Ins 12 Ins 13 Ins 14 Ins 15 Ins 16

Payer

Num

ber

of D

enia

ls

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%n=778


Analyze Root Causes and Initiate Countermeasures • In a half-day root cause analysis session, the team identified ways to change the process

to work around the denials and change the contract process to 1) reduce delays that

contribute to timely filling denials and work with the insurer to resolve excessive denials.

Verify Results: After implementing the process changes the following Monday, denied claims fell by $380,000

per month ($15 million/year). XmR chart below shows denials before and after improvement.


Reducing Rejected Claims In Five Days In software we have a saying that “finding a bug in a computer program is like finding a

cockroach in your hotel room. You don’t say: Oh, there’s a bug. You say: The place is infested.”

The same is true of rejected claims. Start with a line or control chart of rejects:

Use a series of Pareto charts to narrow your focus:


Rejected claims are the frequent type of error; appeals tie up accounts receivable, and denials

result in lost revenue. How can we use Lean Six Sigma? Start with rejected claims.

Categorize Rejected Claims

Rejects by Type

$364,234$893,878$917,622$1,789,756$1,941,177$2,266,610$3,098,585$4,163,173$5,440,543$9,999,612

$19,993,417$20,135,403$20,410,036$20,435,973

$40,418,050

100%99%99%97%96%95%93%90%86%

80%

67%

53%

40%

27%

$-

$19,033,508

$38,067,017

$57,100,525

$76,134,034

$95,167,542

$114,201,050

$133,234,559

$152,268,067

Dup Clai

m

No Cov

erage

Add't I

nfo Req

'd

Provide

r Info

Req'd

Incorr

ect In

s Info D6 D7 D8 D9

D10 D11 D12 D13 D14 D15

Type

Am

ount

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%n=152268067.28

Duplicate claims accounts for 27% of rejected claims. The next four bars of the pareto chart

combined with duplicate claims accounts for 80% of all rejected claims. Each of these five bars

of the pareto chart is an improvement story requiring root cause analysis. Let’s take duplicate

claims down to the next level of pareto chart.


Duplicate Claims Verification

11111115

60

99%97%96%94%93%92%90%

83%

0

9

18

27

36

45

54

63

72

MedicareForward toSecondary

Carrier

PTBAL billed for ptportion after

insurancepmt

billedsecondaryagain after

primarypayment

Denied IC second billbeing sent

for pt portion

UK Unknown

Cla

ims

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%n=72

In this example, secondary payments for Medicare patients accounts for 83% of the duplicate

claims. The team investigated 72 of these secondary payments and found that they had been

paid, but incorrectly coded in accounting. Simple process changes reduced duplicate claims by

$24 million.

Teams Continued With the Other Four “Big Bars” No Coverage turned out to be caused by charges after policy termination (44%).


No Coverage by Code

11233555101421222634

106

202

100%100%99%98%98%97%96%95%92%89%

85%80%

74%

67%

44%

0

57.5

115

172.5

230

287.5

345

402.5

460

CATCIP

COVNCD IP# PE

IAP

RESDEP

LIM O

DEPENDNSC

CBT NFSTU

Cla

ims

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%n=460

DZ Codes CAT Charges after Policy TerminationCIP Cannot Identify PatientCOV CoverageCOV Treatment not CoveredDEP DependentIAP Invalid Alpha PrefixIP# Invalid Policy NumberLIM Limits ExceededNCD No Coverage for DOSNF No FaultNSC No Service CoverageO OtherPE Presumptive Eligibility

Invalid Insurer info led to SSN incorrect and wrong primary insurer (51%).

Invalid Insurer Info

28

8889202223

54

77

89%86%

83%80%

76%

68%

60%

51%

30%

0

32.125

64.25

96.375

128.5

160.625

192.75

224.875

257

SSI OIN VOU CAT CIP NAME ADR NCD PTB OTHER

Code

Cla

ims

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%n=257

SSI - Social Security IncorrectOIN - Other Insurer PrimaryVOU - VoucherCAT Charges after Policy Termination CIP Cannot identify Patient NAME - Correct Pts NameADR - Incorrect Ins AddressNCD - Treatment not Covered PTB - No MCR Part B

Patient Info rejects led to analysis of Other Insurance (41%) and students missing from parent’s

insurance (39%).


Patient Info Required Rejects

336910

5963

98%96%92%

86%

80%

41%

0

19.125

38.25

57.375

76.5

95.625

114.75

133.875

153

Other Insurance Students Other OtherDependents

Babies Auto Demographic

Cla

ims

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

n=153

Results Half-day root cause analysis sessions for each of the “big bars” on these pareto charts and

subsequent improvements resulted in dramatic improvement in “first pass yield” of insurance

claims.

• 72% reduction in ED billing errors

• 60% reduction in impacted charges


Reducing Appealed Claims in Five Days Delayed payments caused by appealed claims can put a hospital in a financial crunch. In 2003,

appealed claims spiked due to Medicare Part B changes. The recent healthcare reform legislation

and subsequent changes will most likely cause further spikes.

Reject Appeals Chart

$15,045,583UCL

CL $8,363,312

LCL $1,681,040$1,681,040

$3,681,040

$5,681,040

$7,681,040

$9,681,040

$11,681,040

$13,681,040

$15,681,040

$17,681,040

10/02 11/02 12/02 01/03 02/03 03/03 04/03 05/03 06/03 07/03 08/03 09/03

Date/Time/Period

Rej

ect A

ppea

ls

Reject AppealsUCL+2 Sigma+1 SigmaAverage-1 Sigma-2 SigmaLCL


Use Pareto Charts to Analyze Appealed Claims: There is a number of ways to analyze appeals data: by patient and appeal type:

Reject Appeals

$19,441$32,456$33,063$243,241$906,352$1,077,005$1,346,122$1,361,097

$13,997,941

$81,343,021

100%100%100%100%99%98%96%95%

81%

$-

$12,544,967

$25,089,935

$37,634,902

$50,179,869

$62,724,837

$75,269,804

$87,814,771

$100,359,739

ED-InPatient

1 2 3 4 5 6 7 8 9

FC

Cha

rges

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%n=100359738.510001

Reject Appeals

$3,256

$8,149,240$12,302,541

$15,341,703

$64,562,998

100%

92%

80%

64%

$-

$12,544,967

$25,089,935

$37,634,902

$50,179,869

$62,724,837

$75,269,804

$87,814,771

$100,359,739

Auth PrecertNotification

Medical Necessity No Cert/Recert for days Timely Filing FC

Type

Am

ount

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%n=100359738.51


From these pareto charts, authorization and pre-certification of admissions from the ED are the

most common and costly appeals. Root cause analysis required team members from the ED and

Time

admissions to identify and reduce Auth/Precert appeals.

Reducing Appealed Claims Cycle

Appeals Delays

0

5000

10000

15000

20000

25000

30000

35000

40000

0 -180

180 -360

360 -540

540 -720

720 -900

900 -1,080

1,080-

1,260

1,260-

1,440

1,440-

1,620

1,620-

1,800

1,800-

1,980

1,980-

2,160

2,160-

2,340

2,340-

2,520

2,520-

2,700

Days

Num

ber

of A

ppea

lsAuth/Precert Appeals by Patient Type

7105266309429

1543

2839

100%98%93%

88%

80%

52%

0

687.25

1374.5

2061.75

2749

3436.25

4123.5

4810.75

5498

In Out S V R E F

Patient Type

Num

ber

of F

B M

emos

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%n=5498


Using the simple tools of Lean (Post-it Notes), it was possible to redesign the appeals process to:

• Reduce touches per account from 21 down to 11

• Minutes per touch (16 minutes)

• 2.97 hours saved per account

• Accelerate payment by 50 days

Other Examples

In 2002, rural hospital, Thibodaux Regional Medical Center, used Lean Six Sigma to reduced

“discharged not final billed” from $3.3 million to $600,000. They also reduced net accounts

receivable from 73 days to 62 days resulting in an increased cash flow of $2 million per year. A

In 2 6 a to reduce oncology

billing errors (missing charges) from 50 percent to only 2.5 percent. This resulted in increased

rev o reduced turnaround time for charge entry from 3.7 to

2.4 days and DOS-to-billing from 13.6 days to 6.8 days. The biggest factor in timely filing of

bills acy:

second wave of projects saved an addition $489,000 per year in inventory costs.

00 , North Shore Long Island Jewish Health System used Six Sigm

enue of $4 million per year. They als

: missing information. Biggest culprit, pharm


How to Get a Cheaper Hospital in Five Days

From working with teams in various industries, I’ve developed a simple method for achieving

breakthrough improvement on transactional processes like billing. I call it the Dirty Thirty

Process. I used it in the case study presented in this white paper.

The Dirty 30 Process for Better Billing

The secret is to:

1. Quantify the cost of correcting these rejected, appealed and denied transactions

2. Understand the pareto pattern of rejected, appealed and denied transactions

3. Analyze 30-50 rejected, appealed or denied transactions to determine the root cause

4. Revise the process and system system to prevent the rejected, appealed or denied claim.

to understand each error. Detailed analysis of 30 errors in each of the top error “buckets” (i.e.,

all

1. Focus: Determine which rejected, appealed or denied error buckets to analyze first for

revent the

3. Sustain: Track the rejected, appealed and denied claims after implementation of the changes.

4. Honor: Recognize and reward team members

Process: Typical root cause analysis simply does not work because of the level of detail required

The Dirty Thirty) led to a breakthrough in understanding of how errors occurred and how to

prevent them. Simple checksheets allowed the root cause to pop out from analysis of this sm

sample. As expected, the errors clustered in a few main categories. The Dirty Thirty process has

four steps:

maximum benefit. (This analysis takes 2 to 3 days.)

2. Improve: Use the Dirty Thirty approach to analyze root causes (4 hours per error type—

facilitator with team) and determine process and system changes necessary to p

problem.


Insights

Using the basic tools of Six Sigma, anyone can learn to use what I call The Dirty Thirty Process

in a day or less to find the root causes of transaction errors. Once a team has found the root

causes of these errors, it’s just a matter of changing the processes and systems to eliminate these

errors.

Hundreds of people spend their lives fixing the fallout from these rejected, appealed and denied

fixing things that shouldn’t be

wrong to begin with.

get to where you can prevent errors, every system could benefit from a simple, yet

the data required to implement it is collected by most systems automatically. Then all it takes is 4

each error.

ance?

en’t

ur, the KnowWare Man, works with hospitals that want to get faster, better and cheaper

claims. And they all think they’re doing meaningful work, not just

Conclusion

Until you

rigorous approach to analyzing and eliminating errors. The Dirty Thirty process is ideal because

to 8 hours of analysis to identify the root cause of

Need Guid

The first project may seem scary, but we can facilitate your improvement teams to achieve

breakthroughs in patient flow. Once you’ve learned how, you’ll find it easy to continue. Hav

you waited long enough to get a faster hospital in five days or less?

Jay Arth

in a matter of days using the proven methods of Lean Six Sigma. Jay is the author of Lean Six

Sigma Demystified and the QI Macros SPC Software for Excel. Jay has worked with healthcare

companies to reduce denied claims by $3 million per year, appealed claim turnaround time and

lab turnaround times by 30-70 percent.

ospital in five days, call: Jay Arthur at 888-468-1537

Email: [email protected]

To get a faster h

Web: www.qimacros.com

Mail: KnowWare, 2253 S. Oneida St. Ste 3D, Denver, CO 80224


Documents

A Cheaper Hospital In Five Days