UHC Webinar Series: Mining the Metrics of Risk Models

Report
™
UHC Webinar Series:
Mining the Metrics of
Risk Models
Obesity and Malnutrition
Palliative Care and DNR Status
May 30, 2012
Presented by Suzanne Rogers, RHIA, CCS, CCDS
1
Agenda
Introduction to the UHC Webinar Series:
“Mining the Metrics of Risk Models”
UHC Risk Adjustment: brief re-cap
Documenting & Coding Series Focus #1
• Obesity, BMI
• Malnutrition, BMI
• End of Life
–
Palliative care
– Hospice
– DNR
DRAFT
04.04.12
2
™
UHC Mining the Metrics of Risk Models Webinar Series
The purpose of this series of webinars will be to highlight some of the
diagnoses & procedures that impact the UHC risk adjustment models or
have the potential to impact these models in the future
Goals & Objectives of the Series:
•
Quarterly webinars with documentation and coding educational focus that will:
−
−
−
−
−
Demonstrate of how documentation and coding practice directly influences the
administrative data used in benchmarking, performance improvement, and
reimbursement
Encourage accurate, compliant, and consistent documentation and coding practice
throughout the UHC membership
Explain the application of Official Coding Rules & Guidelines in a environment of
federal regulations and audits
Examine how current documentation and coding practice of the membership will
impact future UHC risk adjustment models
Measure improved consistency of coded and reported metrics

3
Baseline data will be collected for each focus topic prior to the webinar and then monitored
going forward
™
™
UHC Risk Adjustment
Brief Re-Cap
4
UHC Clinical Database
UHC receives data from members throughout the year. Data is sent with a
minimum of 45 day lag time to allow for the completion of coding and billing cycles
The data feed contains many data elements such as:
•
•
•
•
•
•
ICD-9 diagnosis and procedures codes
Patient demographics such as age, race, sex
Admission source, admission status, and discharge disposition
Encounter physicians and their clinical specialty
Line item charge details
Primary and secondary payers
The MSDRG is not sent; UHC recalculates the MSDRG as well as a APR DRG for
every inpatient encounter
UHC performs data quality checks, flags bad data which does not go to the data
base, and applies the various risk adjustment models
The UHC clinical database risk adjusted data can be used for benchmarking and
evaluating clinical care, patient outcomes and costs
™
UHC Risk Adjustment: Brief Re-Cap
UHC recalibrates the CDB risk models annually
CDB discharges from major academic medical centers are used to build
the models
A coefficient is assigned for each of the variables found to be statistically
significant predictors of the outcome
Only conditions that are Present on Admissions (POA) are considered in
the models
When using the database you can choose to report with the current
(2011) or previous (2010) risk model
DRAFT
04.04.12
6
™
Risk Model Information on Website
From UHC CDB/RM home
page, select Documents &
Presentations drop down list
and choose Risk Methodology
The following screen should appear as
a pop-up window
Choose the Risk Model Summary for
2011
7
8
™
Risk Model Information on Website Cont.
Many other resources that
pertain to the risk models can
also be found here such as:
• Diagnosis & ICD-9 codes that
define the AHRQ
Comorbidities
• ICD 9 diagnosis and
procedures codes used in the
risk models
• Variable definitions for MSDRG
2011 Models
• Definitions and tips on using
the CDB
• Interactive Risk Model
Calculator
9
™
Risk Model Information on Website Cont.
You can use the Interactive Risk Model Calculator to calculate the risk
adjustment for a particular encounter, to determine the impact of deleting
or adding diagnoses or other data elements
10
™
Coding Accurately and Completely
A diagnosis may not be in current risk models but can only be tested for
impact on clinical outcomes if it is coded
•
What is coded today can be important in developing FUTURE risk models
POA is an important aspect of developing and applying the risk models
•
Query if UNKNOWN
• Blank POA = NO for risk adjustment
Demographic data elements are key components of the risk models
•
Patient age, admit source & admit status should be verified
Learn how to access, read, and apply the risk models
•
11
Match the model to the medical record. What got coded? What was missed?
What was the impact? Were the POA’s accurate?
™
™
Obesity, Malnutrition and BMI
12
Obesity, Malnutrition and BMI
Both obesity and malnutrition are factors in AHRQ risk adjustments for
some Quality Indicators and Patient Safety Indicators
Both obesity and malnutrition are factors in UHC’s LOS, cost and
mortality risk models
The ICD-9 diagnosis codes must have a POA of W, Y,1, or E to be
considered in the risk adjustment process
The BMI V-Codes are EXEMPT from POA reporting and should
appropriately be assigned ‘E’ or ‘1’ for POA exempt (do not leave blank)
POA indicators that are BLANK are translated to a N- Not Present on
Admission for risk adjustment
13
™
Obesity and Body Mass Index
Obesity currently a risk factor in 199 UHC risk models (57%)
• Defined by codes:
–
–
–
–
–
–
–
–
–
278.00 Obesity unspecified
278.01 Morbid obesity
649.1X Obesity complicating pregnancy
793.91 Image test inconclusive due to excess body fat
V85.30 Body Mass Index between 30-39, adult
V85.40 Body Mass Index between 40-49, adult
V85.54 Body Mass Index pediatric ≥ 95th percentile for age
Excludes MSDRG 640-641 Nutritional and Metabolic Disorders
Excludes MSDRG 619-621 Operating Room Procedures for Obesity
DRAFT
04.04.12
14
™
Obesity
Diagnosis
UHC Model
AHRQ
Co-Morbidity
CMS – CC/MCC
278.00- Obesity NOS
YES
YES
NO
278.01-Morbid obesity
YES
YES
NO
278.02-Overweight
NO
NO
NO
278.03-Obesity hypoventilation
syndrome
NO
NO
CC
646.1X- Excessive weight gain in
pregnancy
NO
NO
NO
649.1X- Obesity complicating
pregnancy
YES
YES
NO
783.1- Abnormal weight gain
NO
NO
NO
793.91-Image test inclusive; excessive
body fat
YES
YES
NO
V85.30-V85.39- BMI between 30-39,
adult
YES
YES
NO
V85.40-V85.45- BMI between 40-49,
adult
YES
YES
CC
V85.54- BMI, pediatric >95th percentile
for age
YES
YES
NO
DRAFT
04.04.12
15
™
Obesity: Some Facts
A person is considered obese when his or her weight is 20% or more
above normal weight
A person is considered obese if his or her BMI is over 30
"Morbid obesity" means that a person is either 50-100% over normal
weight or has a BMI of 40 or higher
The CDC estimates that 30% of adults are obese and 17% of all direct
medical costs in the US are related to obesity
Being overweight or obese severely interferes with health and normal
function as well as increases risk of morbidity and mortality
Reference: Adult Obesity Facts: Center for Disease Control and Prevention website
DRAFT
04.04.12
16
™
Obesity Coding and CDI Opportunity
What to look for and query for definitive diagnosis
• Body habitus and/or BMI is often documented by nursing as part of an
admission assessment (Joint Commission requires a nutritional assessment
be done at admission)
• You may also find this in the H&P as part of the physical exam or review of
systems
• Is the patient on a low fat diet or limited calories
• Look for a dietary or nutrition consult or note
• Are special measures being taken for imaging, transport or accommodations
• Is there mention of poor imaging results due to body fat
• Does OR report mention any difficulty or prolonged OR time due to the
patient’s size and/or amount of body fat?
DRAFT
04.04.12
17
™
Obesity Coding Guidelines
A code for the BMI can be assigned from nursing/ dietician notes ONLY
if the treating physician documents the associated clinical diagnosis of
obesity, morbid obesity etc.
To be coded the diagnosis of obesity, it must be clinically significant to
the patient’s current hospital stay (it almost always is, so look for any
measures taken or conditions resulting from pts weight)
•
To be coded a secondary diagnosis must affect patient care in terms of
requiring:
−
−
−
−
−
Clinical evaluation
Therapeutic treatment
Diagnostic procedures
Extend LOS
Increased nursing care or monitoring
Reference: AHA Coding Clinic
DRAFT
04.04.12
18
™
Obesity Coding Guidelines Cont.
Obesity Dx must be POA; logically it would have to be since the pt is not
likely to become obese during the hospital stay, HOWEVER the
physician must document it as such (Dx in the H&P, document the
duration, longstanding, etc)
Educate physicians on the importance of documenting the condition of
obesity/morbid obesity and the clinical significance this condition has on
the patient’s care and disease process
19
™
Quarter 4, 2011: All Members
Diagnosis
27800 - obesity nos
POA
Cases
clinically undetermined, provider is unable to clinically determine whether condition was
poa or not
3
no, not present at the time of inpatient admission
39
not reported to uhc
214
present at the time of inpatient admission
58,699
unknown, documentation is insufficient to determine if condition is poa
27801 - morbid obesity
unreported/not used, exempt from poa reporting (1,e)
clinically undetermined, provider is unable to clinically determine whether condition
poa or not
8
4
was
no, not present at the time of inpatient admission
not reported to uhc
present at the time of inpatient admission
6491X - obesity comp
preg
100
34,967
3
unreported/not used, exempt from poa reporting (1,e)
2
no, not present at the time of inpatient admission
8
present at the time of inpatient admission
no, not present at the time of inpatient admission
present at the time of inpatient admission
99% of Obesity Diagnosis are reported with a POA = Y
20
13
unknown, documentation is insufficient to determine if condition is poa
not reported to uhc
79391 - imag inconcexcess fat
1
12
7,223
2
16
Quarter 4, 2011: All Members
Diagnosis
POA
V8530-V8539-BMI 30-39 adult
V8540-V8545 BMI 40 and over adult
V8554 -BMI >=95% for age pedi
Cases
clinically undetermined, provider is unable to clinically determine whether
condition was poa or not
no, not present at the time of inpatient admission
not reported to uhc
present at the time of inpatient admission
unknown, documentation is insufficient to determine if condition is poa
not used, exempt from poa reporting (1,e)
clinically undetermined, provider is unable to clinically determine whether
condition was poa or not
no, not present at the time of inpatient admission
not reported to uhc
present at the time of inpatient admission
5
18,770
unknown, documentation is insufficient to determine if condition is poa
not used, exempt from poa reporting (1,e)
not reported to uhc
present at the time of inpatient admission
1
19,796
185
6
not used, exempt from poa reporting (1,e)
33-36% of adult BMI in obese pts are reported to UHC without a POA indicator
46% of pedi BMI in obese pts are reported to UHC without a POA indicator
Assign POA indicator “1 or E” for Exempt so that the BMI codes can be
used to calculate risk adjustment
(POA’s that are BLANK = “NO” in risk models)
21
1
3
10,770
180
1
9
10,062
284
210
How to Report on Your Obesity Dx POA Assignment
Exclude MSDRG’s
619-621 & 640-641
(per AHRQ definition)
22
™
Create an Advanced Restriction:
AHRQ CC obesity codes and
POA Flag- include all
Save your advanced restriction as a custom list
Click on underlined
numbers to drill down
to case profile
23
Save your report by using the Save icon
In Save Report: popup window give your report a name and file it
under a existing or new group and click save
24
Case #1 DRG 439 Disorder of Pancreas w/ CC
58 yr Black Male
Obsv LOS 12 days / Expected LOS 3.42
Expected Mortality 0.00205
Diagnosis
ICD-9
POA
Mort Variable LOS Variable
Acute
pancreatitis
577.0
Yes- PDX
NA
NA
BMI 50-59.9
V85.43
BLANK (exempt)
0.535
0.1107
Hypoxemia
799.02
YES
NA
NA
OSA
327.23
YES
NA
NA
HTN
401.9
YES
NA
NA
1. BMI present on admission indicator left blank & No Dx of morbid obesity coded.
If added would increase Exp Mort to 0.53705 and Exp LOS to 3.53
2. Combo of morbid obesity, OSA & hypoxemia consider QUERY for pickwickian
syndrome for second CC
25
™
Malnutrition and BMI
Malnutrition currently a risk factor in 224 UHC risk models (64%)
• Defined by Codes:
–
–
–
–
–
–
–
–
–
–
–
260 Kwashiorkor
261 Nutritional marasmus
262 Other severe protein-calorie malnutrition
263.0 Malnutrition of moderate degree
263.1 Malnutrition of mild degree
263.2 Arrested development following protein-calorie malnutrition (nutritional dwarfism)
263.8 Other protein-calorie malnutrition
263.9 Unspecified malnutrition
783.21 Loss of weight
783.22 Underweight
Excludes MSDRG 640-641 Nutritional and Metabolic Disorders
DRAFT
04.04.12
26
™
Malnutrition
Diagnosis
UHC Model
AHRQ Co-morbidity
CMS CC/MCC
260- Kwashiokor
YES
YES
MCC
261- Nutritional marasmus
YES
YES
MCC
262-Other severe protein- calorie malnutrition
YES
YES
MCC
263.0- Moderate malnutrition
YES
YES
NO
263.1- Mild malnutrition
YES
YES
NO
263.2-Arrested development following proteincalorie malnutrition
YES
YES
CC
263.8- Other protein-calorie malnutrition
YES
YES
CC
263.9- Unspecified protein-calorie malnutrition
YES
YES
CC
783.21- Loss of weight
YES
YES
NO
783.22- Underweight
YES
YES
NO
783.41- Failure to thrive, child
NO
NO
NO
783.7- Adult failure to thrive
NO
NO
NO
779.34- Failure to thrive, newborn
NO
NO
YES
V85.0- BMI <19, adult
NO
NO
CC
DRAFT
04.04.12
27
™
Malnutrition: Some Facts
Currently there is no authoritative definition for the diagnosis or severity of
malnutrition …..
May 2012 Academy of Nutrition and Dietetics issued a consensus
statement
Identification & Documentation of Adult Malnutrition (Undernutrition)
• Continuum of characteristics or variables, no one marker for malnutrition
• Inadequate intake, increased requirements, impaired absorption, altered
transport, altered nutrient utilization can cascade into malnutrition
• Inflammation is identified as an important underlying factor
–
Albumin /pre-albumin are indicators of the inflammatory response and not malnutrition
• Two or more of the six characteristics recommended for diagnosis
–
–
–
–
–
–
Insufficient energy intake
Weight loss
Loss of muscle mass
Loss of subcutaneous fat
Localized or generalized fluid accumulation
Diminished functional status (hand grip)
Reference: Journal of the Academy of Nutrition and Dietetics May 2012 Vol.112 number 5
28
™
Figure. Etiology-Based Malnutrition Definitions. Adapted with permission from reference (8): Jensen GL, Bistrian B, Roubenoff R,
Heimburger DC. Malnutrition syndromes: A conundrum vs. continuum. JPEN J Parenter Enteral Nutr. 2009;33(6):710-716.
29
Malnutrition: Some Facts Cont.
Diagnosis is usually based on several factors including some or all of the
following:
• Physical appearance:
–
Cachexia, atrophy, emaciation, unintended or unexpected weight loss
• Risk factors:
–
Cancer, chemotherapy, ETOH, GI or pancreatic disorder, trauma, inflammatory
responses, recent GI surgery
• Test results:
–
Albumin as marker of inflammation, BUN, CBC, protein stores
• None of these tests are markers for diagnosis rather they indicate underlying
etiology for susceptibility
• History:
–
30
Recent or rapid weight loss, decreased functionality, recent trauma or other stress on
patient physiology
™
Malnutrition: Coding and CDI Warning
The terms “severe malnutrition”, “severe calorie deficiency”, “protein
malnutrition”, “emaciation”, “protein deficiency”, “nutritional atrophy”
Code to either:
• Kwashiorkor (260)
• Nutritional marasmus (261)
Both these conditions are highly unlikely to occur in the general
population and are usually associated with children in third world
countries
Both these codes are current targets for review and denial by the OIG &
RAC’s
Assign codes from category 263- Other and unspecified protein-calorie
malnutrition
Educate the physicians on the terminology and code assignment
DRAFT
04.04.12
31
™
Malnutrition: Coding and CDI Opportunity
Look for CLUES and query for clinical significance and/or definitive
diagnosis (must meet the criteria for a reportable secondary diagnosis)
•
•
•
•
BMI <19 in nursing, dietary notes
Patient reported “loss of weight” (783.21) in H&P
Physical exam noted: underweight, cachexic, very thin or frail, failure to thrive
Conditions associated w/malnutrition such as, cancer and cancer treatment,
HIV, s/p GI surgery, depression, prolonged illness and/or hospitalization,
elderly, pancreatic disease, ETOH or drug abuse, swallowing difficulties
Educate the physicians on the coding rules and code assignment
Malnutrition can be coded even if it is the result of or an expected
manifestation of the underlying disease such as cancer or pancreatic
disorders
DRAFT
04.04.12
32
™
Malnutrition: Coding and CDI Opportunity Cont.
Look for CLUES in treatments and orders and query for definitive
diagnosis
• TPN or other enteral feeding
• Ensure, Boost, Carnation Instant Breakfast and other high protein/ high calorie
supplements
• Swallowing studies, GI consult, nutrition consult, OT/PT – ADLs
Look for clinical evidence of malnutrition and/or treatment at admission
A patient rarely becomes malnourished during a routine hospital stay
Educate and query for malnutrition documentation in the H&P,
admit note or designated as being present on admission
DRAFT
04.04.12
33
™
Quarter 4, 2011 All Members
Diagnosis
POA
Cases
260 - kwashiorkor
no, not present at the time of inpatient admission
present at the time of inpatient admission
261 - nutritional marasmus
clinically undetermined, provider is unable to clinically determine whether condition was poa or not
no, not present at the time of inpatient admission
not reported to uhc
present at the time of inpatient admission
unknown, documentation is insufficient to determine if condition is poa
3
529
5
3,130
2
262 - oth severe malnutrition
clinically undetermined, provider is unable to clinically determine whether condition was poa or not
no, not present at the time of inpatient admission
not reported to uhc
present at the time of inpatient admission
unknown, documentation is insufficient to determine if condition is poa
3
754
2
4,169
3
2630 - malnutrition mod degree
clinically undetermined, provider is unable to clinically determine whether condition was poa or not
no, not present at the time of inpatient admission
not reported to uhc
present at the time of inpatient admission
unknown, documentation is insufficient to determine if condition is poa
2
641
1
3,318
5
2631 - malnutrition mild degree
no, not present at the time of inpatient admission
present at the time of inpatient admission
unknown, documentation is insufficient to determine if condition is poa
321
1,745
3
2632 - arrest devel d/t malnut
present at the time of inpatient admission
2638 - protein-cal malnut nec
no, not present at the time of inpatient admission
present at the time of inpatient admission
2639 - protein-cal malnut nos
clinically undetermined, provider is unable to clinically determine whether condition was poa or not
no, not present at the time of inpatient admission
not reported to uhc
present at the time of inpatient admission
unknown, documentation is insufficient to determine if condition is poa
78321 - loss of weight
no, not present at the time of inpatient admission
not reported to uhc
present at the time of inpatient admission
unknown, documentation is insufficient to determine if condition is poa
717
28
9,109
1
78322 - underweight
not reported to uhc
present at the time of inpatient admission
4
1,135
34
16
95
9
119
737
13
5,221
21
16,434
36
In 17% of
the cases
the POA
indicator did
not allow
the DX to
be
considered
for risk
adjustment:
POA= No,
Blank,
Unknown
™
Report on Your Malnutrition Coding Performance
Create an Advanced Restriction: Exclude MSDRG 640
Create an Advanced Restriction: AHRQ CC malnutrition codes & POA flag- include all
(be sure to save your Advanced Restriction as a custom list for future reporting)
35
™
Report on Your Malnutrition Coding Performance Cont.
Click on
underlined
case
numbers to
drill down to
case profiles
In this example 21% of malnutrition codes are indentified as:
POA =No, Not Present on Admission
36
Clarify this diagnosis is 262
Other severe protein-calorie
malnutrition more
appropriate
Look for clues of present on admission
Pt started on enteral nutrition on admission
LOS is only 4 days
Pt has oral cancer
Pt has a trach and gastrostomy
Query for diagnosis based on treatment and pts comorbidties
37
Change “sort by” to
Discharge Phys Specialty
Change Advanced
Restriction to POA =N
Focus on General Medicine:
Perform some chart reviews
to substantiate
38
Provide education for CDI,
coders and Gen Med
physicians based on finding
and examples
™
Palliative Care and DNR
39
Palliative Care and DNR (Do Not Resuscitate)
Palliative care (V66.7) is in 27 UHC risk models
• Exempt from POA reporting
• Always coded as a secondary diagnosis
New code for Do Not Resuscitate (V49.86)
• POA required
• Not in risk models yet, but data is being collected for future consideration
As of July 2012, palliative care will no longer exclude a case from Core
Measures
• Comfort care or DNR must be documented for exclusion
DRAFT
04.04.12
40
™
Palliative Care (V66.7): Some Facts
The National Institutes of Health defines palliative care as relief of
symptoms without curing disease. Hospice care, care at the end of
life, always includes palliative care. But you may receive palliative
care at any stage of a disease. The goal is to make you
comfortable and improve your quality of life.
CMS does not consider palliative care, hospice care, or comfort
care synonymous even though the code assigned for all will be
V66.7
The level and type of care being given must be clearly documented
in the record for Core Measures reporting and hospice
reimbursement
DRAFT
04.04.12
41
™
Palliative Care (V66.7): Some Facts Cont.
Some indications of palliative care could be:
• Withdrawal of any treatment or therapy (e.g. extubation, d/c-ing antibiotics,
vasopressors, other therapeutic drugs
• Addition or increase of opiates and other pain relief treatments
• Treating symptoms, but not the cause/disease (e.g. drainage, relief of
pressure, enteral feeding, oxygen by nasal cannula etc)
• Transfer out of critical care setting, possibly to specialized palliative care
unit/bed or even hospice
Remember palliative care can be given in non-terminal
cases, it is not always end of life
42
™
DNR (V46.89): Some Facts
DNR (do not resuscitate) is a physician order not to perform CPR or
ACLS
• Can also be called a “no –code” or DNR/DNI
Palliative care and DNR/DNI are not the same
Patient can be a DNR and NOT be on palliative care & vice versa
• You can not assume one because the other is documented
Patient can be DNR on admission or can request DNR status anytime
during an admission.
• Living will, advanced directives
DRAFT
04.04.12
43
™
Palliative Care & DNR- Coding & CDI Opportunities
If a patient is on palliative care look for evidence of a DNR, and query
for physician documentation so both can be coded
•
DNR/DNI may be documented in nursing notes, social worker or other care
provider note
• For DNR to be coded it must be documented by a treating physician
If a patient/family decide on withdrawing treatment and/or transfer to
hospice, be sure to have that documented and code the V66.7 if
comfort measures are started prior to transfer
DRAFT
04.04.12
44
™
Palliative Care & DNR- Coding & CDI Opportunities Cont.
Provide general education to physicians on how to document palliative
care, comfort care and DNR as well as the dying process as
appropriate
•
Remember the patient does not need to be terminal to be receiving palliative care or
have a DNR order
When a patient is DNR or on palliative care and they die; look for the
associated conditions of the dying process and query physician to
document
Do not miss out on documentation and coding of CC’s, MCC’s
diagnoses of the dying process
DRAFT
04.04.12
45
™
The Dying Process:
What Might not be Documented/Coded
Diagnosis
Codes
CC
Metabolic disorders
276.0-276.4
X
Encephalopathies
348.30-348.39, 349.82
X
Brain death
348.82
X
Cerebral anoxia/hypoxia
348.1
Cardiac arrest (if d/c to hospice)
427.5
Respiratory failure & insufficiency
518.51-518.53, 518.81518.84
X
X
Acute renal failure
584.5-584.9
X
X
Coma/comatose
780.01
Persistent vegetative state
780.03
X
Cheyne-stokes respirations
786.04
X
Asphyxia/hypoxia
799.01
X
Respiratory arrest
799.1
DRAFT
04.04.12
46
MCC
X
X
X
X
™
Resource
Addendum:
Table of risk models with obesity, malnutrition and palliative care is
included as a hand out for use in identifying impact of these codes on
the various risk models
OBESITY
Risk Model
1
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
47
LOS
Cost
MALNUTRTION
Mortality
LOS
0.203423077
Cost
PALLIATIVE
Mortality
Mortality
0.548716643
0.129202691
1.015232299
3.387205635
0.076908462
2.571205274
0.122894774
0.03836161
0.056922592
-0.100704352
0.062597034
0.136392405
0.187640685
0.149794168 0.10108957
0.1342281 0.057100581
0.194663018
0.471439915
0.260835903 0.18100325
0.310414249 0.178277158
0.390426606 0.42365652
0.223980777 0.151933555
-0.080007655
0.15128321
3.396502738
0.057776625 0.077372606
0.057141883
0.05336678
0.086787678 0.076376272
0.195627179
0.190263684
0.339436109 0.271172886 1.468691594
™
Questions?
Thank You
[email protected]
48

similar documents