HEALTHCARE 1
Sunny Carrington-Hahn
UNCG Honor Code, Academic Integrity Pledge: I have abided by the Academic
Integrity Policy on this assignment.
Signed Sunny Carrington-Hahn Date 10/29/16
ICD-10 ADHERENCE AND QUALITY RESPONSE IN HEALTHCARE 2
Abstract
With the conversion from ICD-9 to ICD-10 coding in healthcare, accurate and
reimbursements. ICD-10 codes that connect care, medications, labs, and procedures to
physicians and other providers. The importance of sound clinical judgement and
prudence with care, medications, labs, and procedures is essential in modern healthcare
that maintains high quality for patients and hospitals. This literature review seeks to
expound on existing research with ICD-10 coding and documentation and its impact on
mortality, productivity
ICD-10 ADHERENCE AND QUALITY RESPONSE IN HEALTHCARE 3
The World Health Organization (WHO) was responsible for adopting the first
monitor and compare mortality statistics and causes of death. Over the years, ICD has
been revised to reflect new knowledge of disease and health. In 1979, WHO released the
8th version of ICD, however, the United States modified ICD-9 by adding more disease
categories and extending coding rubrics to better explain a patients clinical presentation.
This revision of ICD-9 became ICD-9 Clinical Modification (ICD-9-CM), and the most
recent version, ICD-10 used by the U.S., was introduced in 1992 (WHO, 1992).
A foremost distinction between the ICD-10 and ICD-9-CM coding systems is the
difference between the numbers of tabular lists. ICD-10 has 21 categories of diseases
compared with 19 categories in ICD-9-CM and the category of diseases of the nervous
system and sense organs in ICD-9-CM is divided into three categories. ICD-10 includes
diseases of the nervous system, but expands diseases of the eye and adnexa and diseases
of the ear and mastoid process. Also, codes in ICD-10 are alphanumeric while codes in
ICD-9-CM are numeric. For example, each code in ICD-10 starts with a letter (i.e., A-Z),
followed by two numeric digits, a decimal, and a digit (e.g., acute bronchiolitis due to
respiratory syncytial virus is J21.0). Contrary, codes in ICD-9-CM begin with three digit
numbers (i.e., 001999), that are followed by a decimal and up to two digits (e.g., acute
bronchiolitis due to respiratory syncytial virus is 466.11) (Quan, Li, Alibhai, & Ghali,
2008, p. 1425).
innovative its often confronted with realities of fiscal accountabilities that are tied into
ICD-10 ADHERENCE AND QUALITY RESPONSE IN HEALTHCARE 4
quality performance, patient safety, and patient satisfaction. Mostly due to the
complexities of U.S. billing and reimbursement practices, the transition from ICD-9 to
ICD-10 was prolonged and put off for many years compared to other countries as
In addition to intricate billing and repayment systems, healthcare providers in the U.S.
were faced with an enormous clinical documentation culture change. Providers had to
find ways to bridge the gap that exists between the language providers speak to describe a
clinical picture versus language that has to be captured and risk adjusted by ICD-9 and
advisory boards forecasting fiscal impact of ICD-10 varied as it became apparent there
were no clear answer as to how much cash on hand is required to support the ICD-10
transition. Newell & DeSilva (2013) estimated requirements of cash on hand to range
likewise prescribe the days of cash on hand and other required ratios that should be
factored into estimates (p. 78). Healthcare financial leaders were required to pursue
strategies to increase cash on hand in preparation for the ICD-10 transition. For instance,
Newell & DeSilva (2013) explained that a hospital advisory board analysis predicted
hospital was predicted to range from $2.5 million to $7.1 million in lost net revenue,
according to the analysis (p. 79). To clarify, the impact of ICD-9 to ICD-10 was daunting
in many ways to hospitals that the aftermath of the transition highlights the importance of
ICD-10 ADHERENCE AND QUALITY RESPONSE IN HEALTHCARE 5
states for trends in diseases, mortality, and ensuring the most fiscal return for profitable
operations.
Inclusion criteria for this literature review are studies that relate ICD-10 coding to
quality in healthcare or specific diseases. Exclusion criteria for this literature review are
Research Problem/Purpose
The purpose of this literature review is to answer the following questions: (1) how does
ICD-10 coding aid in improving quality in healthcare?, (2) how important is provider
documentation to proper ICD-10 coding?, (3) how does ICD-10 coding affect mortality
data?, and (4) how does ICD-10 coding capture diseases and disease trends?
Methods
Twenty articles were systematically reviewed and analyzed for relevancy to the
finance, and disease classification. Studies were then grouped into three categories:
mortality coding, quality, and disease classification. Databases utilized for article
Findings
Quality
Several articles depicted studies about hospital discharge data and the influence of
ICD-10 coding of quality. Assessing quality of ICD-10 coding by coders through concise
ICD-10 ADHERENCE AND QUALITY RESPONSE IN HEALTHCARE 6
and applicable documentation was important for patient safety, quality, and repayment
motives. Hennessy, Quan, Faris, & Beck (2010) described that the quality of coded data
is influenced by two major factors. The first factor was the clarity, precision, and
which coding was accurate and consistent by providers. The study aimed to find
characteristics and factors by which coding and the coders themselves are influenced to
code with accuracy and reliability. The authors goal was to describe the relationship
between several measures of validity in coded hospital discharge data and define coders'
volume of coding (13,000 vs. <13,000 records), coders' employment status (full- vs.
part-time), hospital type. The study reviewed 422,618 discharge records that were coded
by 59 coders and found that coder characteristics such as experience and training (2-year
data (Hennessy, Quan, Faris, & Beck, 2010). Looking at coder characteristics and their
affect or lack thereof on ICD-10 hospital discharge data, according to Hennessy, et al.,
Diseases, Tenth revision, Canada) had an impact on diagnosis and comorbidity coding in
hospital discharge data. A study was performed by Walker, Hennessy, Johansen, Sambell,
Lix, & Quan, (2012) of nine Canadian provinces that transitioned from ICD-9-CM to
ICD-10 CA over a six year period starting in 2001. The Charlson index was initially
weighted score and then the weighted scores are summed to give an indicator of disease
burden, the Charlson score (Walker, et al., 2012, p. 3). After the conversion of ICD-10-
CA, four provinces found a decrease in the number of diagnoses coded although the
number of diagnoses coded in the other five provinces remained similar. The findings of
the study revealed the implementation of ICD-10-CA in Canadian provinces did not
significantly change coding practices; however some coding variation exists in the mean
number of diagnoses per hospital visit across provinces (Walker, Hennessy, Johansen,
Additionally, a study by Quan, Bing, Saunders, Parsons, Nilsson, Alibhai, & Ghali
(2008), sought to discern whether there were improvements in the validity of coding for
clinical conditions using ICD-10 administrative hospital discharge data compared to ICD-
9-CM coding. The authors reviewed over 4,000 randomly selected charts for patients
admitted during a predetermined time frame in 2003 at four teaching hospitals in Alberta,
Canada to define the presence or absence of 32 clinical conditions and to assess the
agreement between ICD-10 data and chart data. The authors recoded the same charts
using ICD-9-CM to determine consistencies between the ICD-9-CM data and chart data
using the same clinical conditions. The accuracy between ICD-10 data relative to chart
data was compared with the accuracy of ICD-9-CM data relative to chart data, and
overall, 24 out of the 32 clinical conditions were found to have similar sensitivity values
Two studies by Fleming, MacFarlane, Torres, & Duszak (2015) and Stanfill,
Hsieh, Beal, & Fenton (2014) looked explicitly at the impact of ICD-9-CM to ICD-10-
their study how ICD-10 conversion impacted the subspecialty of radiology and its claims
for reimbursement. The findings from this study werent very clear. The authors found
that fewer than 3% of all ICD-9 and ICD-10 codes are used to report the majority of all
radiology claims. Interestingly enough, the number of commonly used ICD-9-CM codes
increased six times with ICD-10-CM codes, with musculoskeletal imaging alone
projected to increase by 30 times. Moreover, Stanfil et al. (2014) pursued to find the
variance in coder productivity using ICD-9-CM versus ICD-10-CM/PCS (PCS stands for
procedure codes) coding classifications, the potential initial productivity loss due to the
coding productivity and quality. The authors determined the average coding of an
inpatient record took 17.71 minutes (69 percent) longer with ICD-10-CM/PCS than with
ICD-9-CM (Stanfil et al., 2014). They inferred through their outcomes that increased
time per case does not necessarily translate to higher quality, and that appropriate training
and subsequent ICD-10 coding can also highlight quality concerns in hospitals. Quan,
Eastwood, Cunningham, Liu, & Flemons, (2013) Strausberg & Hagn, (2015) evaluated
scoring metrics that aided in ICD-10 coding for adverse events in adult acute care
Cunningham, Liu, & Flemons, (2013) looked at discharge data of 490 patients from three
adult acute care hospitals reviewing for documentation on adverse events such as deep
hospitals used the Agency for Healthcare Research and Quality (AHRQ) patient safety
ICD-10 ADHERENCE AND QUALITY RESPONSE IN HEALTHCARE 9
indicators (PSIs) for case findings in the ICD-10 hospital discharge abstract data. The
authors found that ICD-10 coding was a central tool for quality case reporting.
Correspondingly, Strausberg & Hagn, (2015) developed new morbidity and comorbidity
scores founded on ICD-10 coding structure. Routine data from inpatients admitted and
discharged in 2008 from three German hospitals were used for score development, and
same data from 36 German hospitals in 2010 were used for score model evaluation. The
studys appraisal of the morbidity and comorbidity scoring model showed that when used
ICD-10 coding quality can also be ascertained through particular diagnoses. For
ICD-10 colorectal cancer (CRC) diagnosis coding in the Danish National Registry of
Patients (DNRP), using the Danish Cancer Registry (DCR) as a reference was
scrutinized. A large study population of 25, 674 patients who were registered from 2001-
2006 in the DNRP with a CRC diagnosis were included. Data quality was evaluated by
evaluate the effect of potential differences in data quality, and their findings confirmed
the 2004 changes in ICD-10 CRC recording procedures in the DCR, completeness and
PPV were higher in the 20042006 period than in the 20012003 period (Helqvist,
Lastly, two studies by Stausberg, J., Lehmanna, N., Kaczmarekb, D., Stein, M.
(2008) and Southern et. al. (2016) looked at reliability of ICD-10 coding among three
groups of coding subjects and hospital administrative data surfacing ICD-10 codes used
ICD-10 ADHERENCE AND QUALITY RESPONSE IN HEALTHCARE 10
for surveillance of patient safety indicators. Stausberg, J., Lehmanna, N., Kaczmarekb,
D., Stein, M. (2008) enlisted one hundred and eighteen students coded 15 diagnoses lists,
27 medical managers from hospitals coded 34 discharge letters, and 13 coding specialists
coded 12 discharge letters. Members of all three groups agreed on principle diagnoses.
The study concluded that the use of coded data for quality management, health care
financing, and health care policy requires an over simplification of ICD-10 codes to
reflect a true clinical picture. Southern et. al. (2016) employed a sophisticated evaluation
of hospital administrative data and ICD-10 coding to reveal patient safety indicators.
Through a sample of 2,416,413 national hospitalizations, the authors found 2590 unique
ICD-10-CA codes flagged as having arisen after admission. Seven panelists evaluated
these in a 2-round review process, and identified a listing of 640 ICD-10-CA diagnosis
codes judged to be linked to suboptimal quality of care and thus appropriate for inclusion
in PSIs (patient safety indicators) (Southern et. al. 2016, p. 1). The authors determined
Mortality
The following studies from Daking & Dodds (2007), Kim, Shin, Im, Lee, Ko,
Park, Ahn & Song (2009), and Willis, Gabbe, Jolley, Harrison, & Cameron (2010)
discuss the use of ICD-10 in mortality coding and mortality cause discovery. Daking &
Dodds (2007) reviewed the ICD-10 codes provided by the ABS (Australian Bureau of
ICD-10, based only on the information contained in the NCIS (The National Coroners
extracted from 2000-2003 data and comparison to pre-assigned ABS ICD-10 codes. The
ICD-10 ADHERENCE AND QUALITY RESPONSE IN HEALTHCARE 11
two coders were found to be in alignment of each other in 61.1% of cases (Daking &
Dodds, 2007, p. 11). The results of this study show that there is no guarantee that ABS
assigned ICD-10 codes are directly comparable to ICD-10 codes that would be assigned
using the full NCIS record once a coronial case has been completed. In addition, the
study reveals issues in coding external cause deaths before all information is obtainable
Scoring methods in ICD-10 coding with mortality data was developed by authors
Kim et al. (2009), and their study also aimed to validate the method for measuring injury
severity, the excess mortality ratioadjusted Injury Severity Score (EMR-ISS), using the
International Classification of Diseases 10th Edition (ICD-10). The results from the study
revealed EMR-ISS showed better calibration and discrimination power for prediction of
death than the ICISS in most injury groups Kim et al. (2009). The EMR-ISS appears to
be a viable tool for passive injury surveillance of large data sets, such as insurance data
sets or community injury registries containing diagnosis codes (Kim et al. 2009).
Likewise, a study by Willis, Gabbe, Jolley, Harrison, & Cameron (2010) aimed to
compare the performance of ICISS with other mortality prediction tools in an Australian
alternative to mortality prediction tools that use Abbreviated Injury Scale (AIS) data,
including the Trauma and Injury Severity Score (TRISS) (Willis, Gabbe, Jolley, Harrison,
& Cameron, 2010, p. 802). The study was a retrospective review of prospectively
collected data from the Victorian State Trauma Registry, and a training dataset was
created for model development and a validation dataset for evaluation. The
ICD-10 ADHERENCE AND QUALITY RESPONSE IN HEALTHCARE 12
multiplicative ICISS model was compared with a worst injury ICISS approach, Victorian
TRISS (V-TRISS, using local coefficients), maximum AIS severity and a multivariable
affected by the data used to develop estimates, the ICD version employed, the methods
for deriving estimates and the inclusion of covariates. The results of the study showed a
The Charlson comorbidity score for ICD-9-CM adaptation has aided health
Henderson, Perry, Muggivan, Quan, & Ghali (2004) sought to analyze the Charlson
comorbidity score in reference to the Deyo coding algorithm after the ICD-10
conversion. The Deyo coding algorithm was translated from ICD-9-CM into ICD-10-AM
in a layered fashion over time. It was first developed and validated using population-
based hospital data from Victoria, Australia. This work represents the first rigorous
adaptation of the Charlson comorbidity index for use with ICD-10 data. In comparison
in which occupational exposure levels are correlated with year of death. The research
utilized a cohort mortality study in which follow up spans the periods during which ICD-
9 and ICD-10 were in effect. The relative risk estimate obtained when death certificates
are coded to the ICD revision in effect at time of death is compared to the relative risk
ICD-10 ADHERENCE AND QUALITY RESPONSE IN HEALTHCARE 13
estimate that would be obtained if all death certificates were coded to a consistent ICD-10
code. Methods incorporated were simple equations relate the coefficient of bias to the
mortality outcomes that exhibit poor comparability between ICD-9 and -10, it may be
prudent to recode cause of death information to a standard ICD revision in order to avoid
bias that can occur when exposures are correlated with the proportion of deaths coded to
Disease classification
Perhaps the crux of ICD coding is the capability to name and assign patients
clinical presentations for reasons that include improvement of healthcare procedure and
policy, accurate reimbursement, and capturing of disease trends and states. Watzlaf,
Garvin, Moeini, & Anania-Firouzan (2007) and Anderson & Rosenberg (2003) utilized
studies to identify ICD-10-CM coding aptitude for capturing diseases. Watzlaf, Garvin,
Moeini, & Anania-Firouzan (2007) goal of their study was to investigate the
diseases, diseases related to the top 10 causes of death, and diseases related to terrorism)
from users on how applicable the ICD-10-CM systems are in relation to capturing public
health diseases (p. 1). Coder agreements were compared between the AHA (American
Association) ICD-10-CM it was discovered that the ICD-9-CM had higher levels of
agreement than ICD-10-CM, however, general results validate that ICD-10-CM is more
ICD-10 ADHERENCE AND QUALITY RESPONSE IN HEALTHCARE 14
precise and completely captures more of the public health diseases than ICD-9-CM.
Furthermore, in a study by Anderson & Rosenberg (2003) ICD-10 coding for some
leading causes (for instance, septicemia, influenza and pneumonia, Alzheimers disease,
and nephritis, nephrotic syndrome and nephrosis) the incoherence in disease data trend is
coding wasnt limited to studying general disease capturing. A study by Kokotailo & Hill
coding versus ICD-9 coding through hospital administrative data from 2000-2003 in
Canada. The authors founds that stroke coding was equally good with ICD-9 (90%
[CI95 86 to 93] correct) and ICD-10 [92% (CI95 88 to 95 correct) with ICD-10. There
were some differences in coding by stroke type, notably with transient ischemic attack,
but these differences were not statistically significant. Likewise, atrial fibrillation,
coronary artery disease/ischemic heart disease, diabetes mellitus, and hypertension were
coded with high sensitivity (81% to 91%) and specificity (83% to 100%). They
concluded ICD-10 was as good as ICD-9 for stroke risk factor coding (Kokotailo & Hill,
2005, p. 1776).
In another study from Germany, authors Strausberg & Hasford (2010) evaluated
potential usefulness of ICD-10 coded diagnoses in routine hospital data for the
identification of ADE. Four sources were utilized to identify symptoms and diseases that
can occur as ADE. The data revealed 505 ICD-10-GM 2009 codes that indicate
suspected ADE, only 0.7% of hospital admissions were revealed by routine data to be
causally related to the administration of a drug, and in 5.3% of admissions there was at
least a reason to suspect relationship contributed to drug administration and ADE. The
ICD-10 ADHERENCE AND QUALITY RESPONSE IN HEALTHCARE 15
study uncovered that ICD-10 coding was used effectively and that its practicality for the
identification of ADE should be better exploited (Strausberg & Hasford, 2010, p. 23).
Furthermore, usefulness for ICD-10 coding is the ability to capture accidents that
can lead to disease states and conditions. Karkhaneh & Voaklander (2012) inspected the
(external cause of injury codes) in ICD-9 and ICD-10. They studied 180 randomly
selected bicycle and pedestrian injury charts from Canada hospitals nationwide, reviewed
them, and used the data as the reference standard (RS). To clarify, bicycle and pedestrian
injuries were identified from the ED information system from one period before and two
periods after transition from ICD-9 to -10 coding. The study concluded that bicycle
injuries are coded in a reliable and valid manner; however, pedestrian injuries are often
The quality of the studies used in this review was of high caliber. Study
populations sizes were mostly large with ranges of hundreds to millions. Data in these
studies were able to ascertain conclusions with high confidence. Also, study focus
questions were adequately conveyed and described. Literature search strategy was
Interpretation of Results
Results from this literature review answered the initial questions of: (1) how does
ICD-10 coding aid in improving quality in healthcare?, (2) how important is provider
documentation to proper ICD-10 coding?, (3) how does ICD-10 coding affect mortality
data?, and (4) how does ICD-10 coding capture diseases and disease trends? Overall, it
ICD-10 ADHERENCE AND QUALITY RESPONSE IN HEALTHCARE 16
was found that ICD-10 coding allows patient data to be captured more systematically and
presents clinical information more accurately than ICD-9 coding. Disease and mortality
ICD-10 coding was just as consistent, reliable, and precise compared to ICD-9 coding.
ICD-10 coding has more variability, which allow providers flexibility to reflect complex
Discussion
With the U.S. implementing ICD-10 coding and its prominence in healthcare,
local and global information sharing regarding quality, mortality, and disease is realized
with enormous potential for treatment and research progression. The studies that
discussed newer ICD-10 implementation did so with a tone of caution and reservation
more in U.S. based articles as expressed by Newell & DeSilva (2013), clinical
improved coding efforts all can contribute to accelerating net revenue and cash on hand.
from a bank during the transition period, communicating with the bank how the
organization will manage the fiscal impact of the transition to ICD-10 (p.78-79). ICD-
coding is critical for medication coverage and reimbursement for services rendered.
Quality improvement with procedures, policies, and billing practices should leverage the
References
Anderson, R.N., Rosenberg, H.M. (2003). Disease classification: measuring the effect of
10.1002/sim.1511
Daking, L., Dodds, L. (2007). ICD10 mortality coding and the NCIS: a comparative
Green, D. (2014). New ICD-10 coding: Documentation to provide better care, support
Helqvist, L., Gammelager, H., Johansen, M.B., & Sorensen, H.T. (2012). Quality of ICD-
2354.2012.01350.x
Hennessy, D.A., Quan, H., Faris, P.D., Beck, C.A. (2010). Do coder characteristics
Fleming, M., MacFarlane, D., Torres, W.E., Duszak, R., 2015. Magnitude of Impact,
ICD-10 ADHERENCE AND QUALITY RESPONSE IN HEALTHCARE 19
http://dx.doi.org.libproxy.uncg.edu/10.1016/j.jacr.2015.06.014
journal of the International Society for Child and Adolescent Injury Prevention,
Kim, J., Shin, S.D., Im, T.H., Lee, K.J., Ko, S.B., Park, J.O.,Song, J.K. (2009).
2712.2009.00412.x
Kokotailo, R.A., Hill, M.D. (2005). Coding of stroke and stroke risk factors using
doi: 10.1161/01.STR.0000174293.17959.a1
ICD-10 ADHERENCE AND QUALITY RESPONSE IN HEALTHCARE 20
Newell, L.M., DeSilva, J.J. (2013). Dollars and sense: mitigating budget risk for ICD-10:
there are 10 challenges hospitals and health systems should consider as they
budget resources for ICD-10 and prepare for the financial impact of
Quan, H., Bing, L., Saunders, L.D., Parsons, G.A., Nilsson, C.I., Alibhai, A., & Ghali,
6773.2007.00822.x
Quan, H., Eastwood, C., Cunningham, C.T., Liu, M., Flemons, W., (2013). Validity of
AHRQ patient safety indicators derived from ICD-10 hospital discharge abstract
data (chart review study). BMJ OPEN, 3(10): 1-7. doi: 10.1136/bmjopen-2013-
003716
Southern, D.A., Burnand, B., Droesler, S.E., Flemons, W., Forster, A.J., Gurevich, Y.,
Harrison,N., Ghali, W.A. (2016). Deriving ICD-10 codes for patient safety
Stanfill, M. H., Hsieh, K.L., Beal, K., & Fenton, S. (2014). Preparing for ICD-10-
Strausberg, J., Hagn, S. (2015). New morbidity and comorbidity scores based on the
doi:10.1371/journal.pone.0143365
Strausberg, J., Hasford, J. (2010). Identification of adverse drug events the use of
ICD-10 coded diagnoses in routine hospital data. Dtsch Arztebl Int. 107(3): 23
Stausberg, J., Lehmanna, N., Kaczmarekb, D., Stein, M. (2008). Reliability of diagnoses
doi:10.1016/j.ijmedinf.2006.11.005
Sundararajan, V., Henderson, T., Perry, C., Muggivan, A., Quan, H., Ghali, W.A. (2004).
10.1016/j.jclinepi.2004.03.012
Walker, R.L., Hennessy, D.A., Johansen, H., Sambell, C., Lix, L., Quan, H. (2012).
doi: 10.1186/1472-6963-12-149
Watzlaf, V, Garvin, J.H., Moeini, S., Anania-Firouzan (2007). The effectiveness of ICD-
Willis, C., Gabbe, B., Jolley, D., Harrison, J.E., Cameron, P.A. (2010). Predicting trauma
2197.2010.05432.x
disease and related health problems, tenth revision (ICD-10). Geneva: World