Content uploaded by Erdi Dasdemir
Author content
All content in this area was uploaded by Erdi Dasdemir on Feb 06, 2017
Content may be subject to copyright.
European Network for Business and Industrial Statistics (ENBIS) 13 Conference
Ankara / Turkey, 15-19 September 2013
1
IMPROVING HOSPITAL BILLING PROCESSES FOR REDUCING COSTS OF
BILLING ERRORS
Erdi DASDEMIR1, Macit Mete OGUZ1, Murat ATALAY1, Volkan BILGIN1,
Murat Caner TESTIK (Ph.D) 1, Guray SOYDAN(M.D, Ph.D) 2
1 Hacettepe University, Department of Industrial Engineering, Ankara, TURKEY
2 Hacettepe University, Faculty of Medicine, Department of Pharmacology Ankara, TURKEY
ABSTRACT
Hospital billing process is a crucial component for hospital management. Due to the
complexity of the hospital billing processes, billing errors may result in costly financial
losses. In Turkish social security system, Social Security Institution (SSI) provides health
insurance, which ensures maintenance of health statuses of individuals and the financing
of costs that arise in case the individuals experience health risks. Accordingly, SSI
developed billing procedures for hospitals in financing the healthcare needs of
individuals. Hospitals need to comply with the standards set by SSI in order to prevent
stoppages and fines in financing their costs. In the following, Hacettepe University
Hospitals’, where 95 % of the healthcare service payments are made by the SSI, are
studied. Nevertheless, there is a huge amount of financial losses from SSI because of the
errors occurring during billing process. Here, the aim is to minimize Hacettepe University
Hospitals’ billing errors. To realize this aim, Lean Six Sigma framework and problem
solving methods of DMAIC are used. The billing process of the hospital is studied first
and critical points are determined. After meetings with the hospital IT personnel and
hospital administration, some important data, including the past billing errors are
retrieved. The main billing errors, their reasons and the financial costs of the errors are
analyzed with statistical and graphical tools. To solve the problems and remove the
errors, work flow and standard operating procedures of the hospital billing process is
prepared.
Key Words: Six Sigma, Process Improvement, Lean Hospital, Medical Billing Process,
Billing Errors
European Network for Business and Industrial Statistics (ENBIS) 13 Conference
Ankara / Turkey, 15-19 September 2013
2
1. Introduction
Hospital medical billing is the process of
submitting and following up on claims to
insurance companies in order to receive
payment for services rendered by
hospitals. The hospital billing process
begins when a patient arrives to the
hospital for diagnosis and treatment of
an injury, illness, disease, or condition.
Until the patient’s departure, all of the
services and items provided are recorded
to patient’s account. In the next step, all
information and charges are processed
for billing based on the requirements of
insurance companies, which generally
have reimbursement rules. During the
billing process, if a hospital does not
meet the requirements of an insurance
company, stoppages to hospital
payments and fines to the hospitals are
generally charged.
In Turkey, Social Security Institution
(SSI) is the major healthcare insurance
provider for individuals and thus it is the
most important organization for
financing healthcare costs. To manage
the costs of individuals’ healthcare needs
with the healthcare providers
participating in the system, SSI
established some billing procedures
called SUT rules. These procedures set
restrictions to the payments to the
hospitals based on some standards.
Billing errors are determined through
sampling of the bills and then stoppages
and fines are charged to the hospitals.
Therefore, mistake proofing in medical
billing process became a crucial
component for hospital management.
In this research, Hacettepe University
Hospitals, where 95 % of the healthcare
services payments’ are made by the SSI,
are considered. This very high
percentage of payments through SSI is
an indicator of the importance of SSI for
Hacettepe Hospitals. Nevertheless, due
to many errors occurring during billing
process, stoppages and fines charged by
the SSI significantly degrades the
financial performance of the Hacettepe
Hospitals. To minimize financial losses
caused by the billing errors, Lean Six
Sigma framework and problem solving
methodology of DMAIC are used in the
following.
2. Literature Review
Although process improvement studies
are popular in manufacturing industries,
interest to their applications also
increased in healthcare industries
recently. There are many application
areas of process improvement in
healthcare; such as flow of patients in
hospitals, medication processes,
medication quality improvement, billing
processes and maintenance processes.
Using lean manufacturing methods, Ben-
Tovim et al. (2007) conducted a study to
improve flow of patients and simplify
complex processes in hospitals. In this
study, journey of patient in emergency
department from arrival to departure are
considered as transforming a raw
material to a finished product. A patient
flow diagram of patients in the
emergency department was created and a
triage score procedure was developed.
European Network for Business and Industrial Statistics (ENBIS) 13 Conference
Ankara / Turkey, 15-19 September 2013
3
As a result, 48 minutes reduction in
average time patients spent in the
department was achieved.
In another study, Koning et al. (2006)
aims to produce systematic innovation
efforts in healthcare, controlling cost
increases and improvement in quality by
using lean and six sigma methodologies
together. In the Red Cross Hospital in
Netherlands, errors in bills received from
temporary agencies were determined.
Value-added steam mapping was created
and significant non-value added rework
and unnecessary re-administrations were
detected. For this problem, critical-to-
quality characteristic was defined as
percentage of correct bills received from
temporary agencies. The root causes of
this problem were then determined and it
was observed that main cause was the
use of different worksheets at
departments. As a solution, worksheets
were standardized and a visual
management system was introduced. As
a result, rework is reduced and cost
savings are achieved. As another
problem, maintenance of mechanical
breakdowns is also investigated in this
study. This time, critical-to-quality
characteristic is defined as repair time of
breakdowns. After analysis step, it was
observed that maintenance departments
did not have standard operating
procedures (SOPs). For this department,
SOPs were prepared, a work planning
system performance indicator and a
visual management system were
introduced. These simple systems
decreased repair time and increased
financial savings.
Adams et al. (2002) conducted a study to
reduce medical billing errors by
monitoring and auditing medical record
documentations. It was emphasized that
loss of revenue to physicians due to
billing errors may be important. Possible
causes of billing errors and risk
assessments of loss of revenue were
determined and corrective action plans
were determined to improve medical
coding and billing practices.
Information management systems are
very important for detecting and
reducing billing errors. Malone (2006)
developed a Hospital Payment
Monitoring Program to reduce billing
errors caused by billing outpatients as
inpatient. This study emphases the
importance of the collaboration of
hospital information management with
case management registration and
billing. As a result of this study, 63.63 %
reduction in billing errors was reported.
3. Methodology
In this study, Lean Six Sigma
methodology is used for improving
billing process through reduction of
errors. DMAIC approach is used in the
project management.
In order to understand and define the
problem, initial investigations through
observation of the billing process and
discussions with experts were done.
These studies and consequences of
errors in terms of costs to the hospitals
indicated that redesign of billing
processes is crucial. In the measure step,
to evaluate and understand the current
state of the system, data that are related
European Network for Business and Industrial Statistics (ENBIS) 13 Conference
Ankara / Turkey, 15-19 September 2013
4
to billing errors were collected. Main
objective was to collect data on the costs
to the hospitals due to billing errors,
frequencies of the errors and types of
errors. As an objective of the analyze
step, root causes of billing errors were
identified. Because a redesign of a
billing process was aimed, processes
relevant to billing process were
investigated. In the analyses,
Exploratory Data Analysis tools were
used. Work flow charts were prepared to
visualize the billing process. Frequency
of errors was pointed in defect
concentration diagrams. The most
critical error types were determined
based on their costs to the hospital.
Process flow charts and standard
operating procedures (SOPs) of
responsible units were prepared.
Moreover, a classification tree was
constructed to determine effects of
patient type and medical specialty type
on the billing errors. The knowledge
discovered from the data analyses
indicated root causes of problems as
well as their locations in the process,
which were later used for process
improvement studies. In the
improvement step, current status of the
system was improved by eliminating the
most critical medical billing errors. The
deficiencies of the current billing
processes were improved and new job
descriptions were defined. During the
improvement studies, the main
philosophy was the Poka Yoke
approach. It was tried to eliminate
billing errors by applying not only
manual preventions but also hospital
automation improvements. With the
software additions to the current
automation system, the errors which
occurred due to personnel failures were
reduced. The study is concluded without
a control step because SSI has a 6 month
lag for examining the medical bills of
hospital. It is planned to observe the
outcomes and financial results in order
to evaluate the achievements.
4. Results
The following subsections provide
details on the results of analyze and
improve steps of the project. Recall that
the main purpose of the project was to
decrease medical billing errors and
hence improve the bottom-line results of
the hospitals by reducing fines and
stoppages from SSI.
In the study, data were gathered from the
Information Technology Department of
Hacettepe Hospitals. Our dataset
consisted of related records from
January to March 2012, since they were
the latest months examined by SSI when
this project was started. To analyze the
data, preprocessing was conducted first.
The dataset, which includes 48000
transactions, had nearly 2500 different
error types. The types of errors in the
data were categorized and decreased
from 2500 to 850. Statistical tools such
as histograms and Pareto charts were
used to visualize the results to make the
conclusions cleaner.
4.1 Data Mining Study
To understand the effect of patient type
and medical specialty type on the costs
of errors, a classification tree shown in
figure 1 was constructed by using SPSS
European Network for Business and Industrial Statistics (ENBIS) 13 Conference
Ankara / Turkey, 15-19 September 2013
5
Clementine 11.1. Here, patient type and
medical specialty type were determined
as nominal inputs, stoppage cost
(categorized based on magnitude as low,
medium, high and extreme) was
determined as the ordinal output.
Following this, a categorization of 8
critical billing errors was done. This 8
critical error types with respect to their
costs are given in Table 1. The process
related to these errors was then
identified through meetings with the
hospital administration and experts.
Causes of the errors were investigated in
the analysis step using flowcharts and
through analysis of current standard
operating procedures.
4.2 Process Analysis Studies
In order to find the error locations in the
process and eliminate these, first the
current process is analyzed. Flow
charting is found to be an important
graphical tool for this study because it
may help showing the errors’ locations
and associate errors with them. Three
different processes;
General Medical Billing Process
Billing Control Process
After Sampling Invoice and
Relevant Document Preparation
Process
were determined as the most important
components of medical billing process
and their work flowcharts were drawn.
A screenshot from the flowchart of
general medical billing process is given
in the following Figure 2.
In addition to the flow charts, standard
operating procedures were designed and
presented in order to define job
descriptions for the following
responsible units: Secretaries (polyclinic
services, patient admission, clinical
services), IT, Pharmacy, Nurseries,
Technicians, Doctors, Laboratories,
Imaging Methods, Radiological, Nuclear
Medicine, Operation Rooms,
Procurement Department, Invoice
Department, Barcode Department.
4.3 Root Causes of the Critical Billing
Errors
Note that 8 critical errors were
determined. After the completion of
process analysis, root causes and
locations of these critical billing errors
were determined. Here;
Material-Barcode-KIK Error is the kind
of error, which is caused by the lack of
or the inappropriate invoices of medical
materials used, by the missing barcode
or by missing required KIK (Public
Procurement Authority records)
documentation.
Insufficient Epicrisis Error is the kind of
billing error, which occurs because of
missing result reports or the
inappropriate result reports.
Duplication Error is the kind of billing
error, which occurs when the same
treatment is reapplied to the patient
without any explanation.
Medicine Error is the kind of error,
which occurs because of a medication
that is not in the rules. Note that
especially in clinical services, there are
several SUT rules about the medicines.
European Network for Business and Industrial Statistics (ENBIS) 13 Conference
Ankara / Turkey, 15-19 September 2013
6
Some kind of medicines can be used
only for specific cases.
Missing Result Error is a kind of error,
which occurs if the results of
examinations are missing. SSI demands
results of the examinations to verify that
they are implemented.
Mismatch of Diagnosis and Tests Error
is a kind of error, which is caused by the
wrong, missing or insufficient diagnosis
and the inconsistency of examinations
and diagnosis.
Judicial Report Error is a kind of error,
which occurs in the emergency
department. Emergency department has
to collect necessary legal reports to
prove the situation.
Per Case Error is the term that is used
to define each type of specific treatments
for different departments and different
patients.
4.4 Classification of Billing Errors
with Respect to Responsibilities
Errors were associated with responsible
by using the general medical billing
process. By identifying the locations and
responsible of errors, it became easier to
find and present accurate improvements.
Table 2 shows classified errors with
respect to responsible.
4.5 Solution Proposals for
Improvement
Material-Barcode-KIK Error
Unifying Procurement and Barcode
Departments under a single department:
This error was caused from
communication problems between the
Procurement and Barcode Departments.
Lack of medical material bills, barcodes
and KIK prints were the main reasons of
the error. Hence, unification of these two
departments may simplify information
flow and reduce errors.
Computerized Archiving:
When material invoices are collected,
transferring of the related invoices to
hospital automation system with certain
codes may be helpful for accessing and
collecting documents which are
demanded by SSI.
Insufficient Epicrisis Error
Training for SUT Rules:
Training of doctors, secretaries and
responsible staff will decrease mistakes
related to lack of knowledge.
Pre-Control Mechanisms:
Doctors are making some mistakes in
submitting the information because of
their high work-load. For that reason,
medical secretaries can control the
information submitted by doctors and
clarify that epicrisis report is convenient
by systematic approval.
Duplication Error
Improving Hospital Automation:
When doctors demand same treatment
twice, they should be warned by a
signal, which remarks the explanation
requirement for the second treatment.
Doctors should not pass to the next stage
without an explanation.
European Network for Business and Industrial Statistics (ENBIS) 13 Conference
Ankara / Turkey, 15-19 September 2013
7
Medicine Error
Improving Hospital Automation:
To eliminate this error, automation
system should be adjusted such that the
system should warn user about the
specific medicines. Also, it should
require an approval to use these specific
medicines.
Per Case Error
Improving Hospital Automation:
Treatment packages must be submitted
to hospital automation system by the IT
department. After diagnosis and
examination, doctors must select a
relevant treatment package.
Judicial Report Error
New Staff:
Besides emergency secretary staff, there
must be specific personnel that are
responsible for collecting and reporting
judicial documents.
Improving Hospital Automation:
If hospital automation does not include
the judicial case report, the system itself
must disable sending of patient invoice
to SSI via Medula system.
Mismatch of Diagnosis and Tests
Improving Hospital Automation:
If doctors demand diagnosis,
examinations or screenings outside their
specialization, hospital automation
system must signal a warning, which
must explain that transaction
inappropriate. If this demand is really a
requirement for the patient, doctors must
submit relevant explanation in order to
support this demand.
Medical Secretaries Department:
Medical secretaries department should
be established to control the consistency
between diagnosis and transactions. This
department’s staff should observe
consistency of SUT rules with diagnosis
and transactions. If there are some
mistakes, doctors should be warned to
correct these mistakes.
Missing Result
Improving Hospital Automation:
A newly designed system can eliminate
this error by preventing the doctors and
nurses to continue their transactions
without uploading relevant reports to
hospital automation system.
Medical Secretaries Department:
Another suggestion for this error is that
medical secretaries should control
diagnosis, examinations and screening
results. If there is insufficiency in these
results, doctors and nurses should be
warned.
4.6 Results of Applied Improvements
After presenting the solutions above for
process improvement, most were
applied. Expected financial outcomes of
these solutions are shown in Table 3.
5. Conclusions
The primary goal of this project was to
improve medical billing process and
hence reduce stoppage costs due to
billing errors at Hacettepe University
Hospitals. Lean Six Sigma framework
and problem solving methods of
European Network for Business and Industrial Statistics (ENBIS) 13 Conference
Ankara / Turkey, 15-19 September 2013
8
statistical quality control are used to
identify the root causes of billing errors.
Billing errors and their financial costs
are analyzed with statistical and
graphical tools. To solve the problems
and remove the errors, work flow and
standard operating procedures of the
hospital billing process is prepared.
Through analysis of the billing
processes, some improvements to
prevent errors are recommended.
6. Acknowledgements
This research is supported by The
Scientific and Technological Research
Council of Turkey (TUBITAK) under
2209 program and Hacettepe
University Scientific Research
Projects fund.
7. References
Adams DL., Norman H., J. Burroughs V.
“Addressing Medical Coding and Billing
Part II: A Strategy for Achieving
Compliance A Risk Management
Approach for Reducing Coding and
Billing Errors”, Journal of The National
Medical Association Vol. 94, No. 6
(2002).
Koning, H., Verver JPS., Heuvel J.,
Bisgard S., Does R.J.M., “Lean Six
Sigma in Healthcare”, Journal for
Healthcare Quality, Vol:28, No:2, .4-11
(2006).
Malone SM., Billing Error Reduction
Project: A Hospital Payment Monitoring
Program Special Study, 2006, Accessed
date: 10.11.2012, Available from:
http://www.cfmc.org/files/review/revie
_CM_AHIMA%20Article.pdf.
SGK SUT rules (Fatura İnceleme Usul
ve Esasları); Accesed date: 01.06.2013;
Available at www.sgk.gov.tr. Tovim,
DB., Bassham, JE., Bolch, D., Martin,
MA., Dougherty, M. and Szwarcbord,
M. “Lean thinking across a hospital:
redesigning care at the Flinders Medical
Centre”, Australian Health Review,
Vol:31, No:1 (2007).
European Network for Business and Industrial Statistics (ENBIS) 13 Conference
Ankara / Turkey, 15-19 September 2013
9
Figure 1. Classification Tree for Categorized Stoppage Costs
Table 1: Most Critical Error Types with Respect to Stoppage Cost (January-March 2012)
Error Types
Error Numbers
Error Percentage to Total %
Stoppage Cost (TL)
Judicial Report Error
2312
14
99.131,14 TL
Medicine Errors
1051
7
56.507,74 TL
Material-Barcode-KIK
Errors
3694
23
408.402,42 TL
Duplication Errors
1575
10
50.884,96 TL
Missing Result Errors
1808
11
244.491,07 TL
Mismatch of Diagnosis and
Tests Errors
2987
19
64.572,92 TL
Per Case Errors
1437
9
42.480,66 TL
Insufficient Epicrisis Errors
1141
7
41.629,68 TL
Grand Total
16005
100
1.008.100,59 TL
European Network for Business and Industrial Statistics (ENBIS) 13 Conference
Ankara / Turkey, 15-19 September 2013
10
Figure 2. A Screen Shot from General Medical Billing Process
Table 2: Classification of Billing Errors Based on Responsibilities
Medical Billing Error Names
Responsible Unit
Material-Barcode-KIK
Procurement Department
Insufficient Epicrisis
Doctors
Nurses
Secretaries
Duplication Error
Doctors
Nurses
Secretaries
Medicine Error
Pharmacy
Missing Result
Secretaries,
Doctors.
Mismatch of Diagnosis and Tests
Doctors,
Secretaries
Judicial Report Error
Emergency Dep.
Per Case Errors
Doctors
Nurses
Invoice Dep.
European Network for Business and Industrial Statistics (ENBIS) 13 Conference
Ankara / Turkey, 15-19 September 2013
11
Table 3: Expected Outcome of Improvements per Month
Error Types
Expected Outcome (TL) per
month
Judicial Report Error
33.043,71 TL
Medicine Error
18.835,91 TL
Material-Barcode-KIK Error
136.134,14 TL
Duplication Error
16.961,65 TL
Missing Result
81.497,02 TL
Mismatch of Diagnosis and
Tests
64.572,92 TL
Per Case Error
14.160,22 TL
Insufficient Epicrisis Error
13.876,56 TL
Grand Total
379.082,14 TL