Indicators of Patient Safety: The Applicability of Veteran Affairs Discharge Data

This study explores the use of Veteran Affairs (VA) discharge data in measuring patient safety in healthcare centres. The study focuses on the development and applicability of the VA discharge data software and compares the performance of hospitals that use the software and those that do not use it. The study concludes that the VA database software is advantageous in that it accommodates both acute and nonacute care hospitalisations.

Running head: PLAY THERAPY 1
Play Therapy
Student’s Name
Institution Affiliation

INDICATORS OF PATIENT SAFETY 2
PLAY THERAPY
Introduction
There are various adverse experiences that patients encounter when they seek medical
services in a healthcare centre. It should be noted that the said complications are different
from the main reason why the patient visited the hospital. Some of these complications
originate from the status of the health facility itself where the patient is admitted (Alyousef et
al., 2017). The Patient Safety Indicators (PSIs) are a set of measures taken to show some of
these adverse experiences the patients have due to exposure to the systems within the
healthcare centre. These adverse experiences can be prevented by merely improving the
conditions that promote them or changing the system in the health facility. The definition of
Patient safety indicators can be grouped into two distinct categories namely provider level
indicators and area level indicators. Provider level indicators enable measurement of the
avoidable adverse experiences by the patients who received their initial medical care and
experienced the complications in the same hospital. Those indicators involve the only case
where a secondary diagnosis code causes another avoidable difficulty (Carls, Henke, Karaca,
Marder, & Wong 2015). Area level indicators showcases of such experiences associated with
a given area like a country or a town. In this type of patient safety indicators, even cases of a
patient suffering some complications after being readmitted in another hospital within a given
area are included.
It should be noted that previously, the measures that were used to evaluate such
incidences were confined to capture cases in the inpatient's settings only. However, the
current PSIs possess a step in development by including all other areas in the hospital and
even in a given geographical area according to Clark, Weinreb, Flahive, and Seifer (2018).
This paper focuses on globally accepted PSI developed through the dynamic four-step

INDICATORS OF PATIENT SAFETY 3
process of literature review, evaluation by clinical practitioners, codes reviewed by industry
experts, and finally subjected to empirical analyses to ensure applicability. This study focuses
on how the use of Veteran Affairs (VA) discharge data improves the patients' safety as
compared to cases where VA discharge data is not applied (Cohen, Cohen, Stagnitti, &
Lefkowitz 2016). The study concentrates on the use of version 2.1 of the software.
Advantages of VA Discharge Data
a. Applicable in cases of large population hence facilitates population-level assessments
based on a calculation of event rates
b. Relatively cheaper as compared to other related software
c. The software focuses on the unique attributes of hospital discharges
d. The software is readily available
e. VA discharge data software is readable in most computers.
The development and application of VA discharge data have enabled the managing of
potential threats to patients’ safety in hospitals and the user as a benchmark for hospital
performance. The use of this technique has however raised concerns that it increased
mortality, prolonged stay in the hospital and increased hospital charges. The importance and
the roles played by the software have however surpassed the concerns raised.
The primary objective of this paper is to show the development and applicability of
the Veteran Affairs (VA) discharge data and compare the performance of hospitals that use
the software and those that do not use it (Dorflinger et al., 2014). The software was
developed and tested using computerised hospital discharge data, and therefore the
definitions used here are based on variables from standard hospital discharge abstracts. The
abstracts used are generated from the clinical and nonclinical data elements that are regarded
as institutional claim standard. It is worth noting that VA databases are developed using

INDICATORS OF PATIENT SAFETY 4
unique formatting system that exhibits both acute and no acute care. This is entirely different
from other widely used hospital administrative databases that only contain standard discharge
abstracts (Helm et al., 2016). The paper intends to generate a valid indicator rate utilising the
software and draw a comparison between the frequencies used in VA and those hospitals that
do not apply VA discharge database. Another target of this study is to present the current
knowledge to future researchers and practitioners who may want to use the information in
future.
Applicability of VA Discharge Database
Some simple information worth knowing regarding this software is that Veteran
Affairs has an administrative database that consists of information on diagnosis, demography,
and how the resources were utilised on all the veterans who visited healthcare centres in VA.
The primary variable being analysed here is the hospitalisation, which is linked to the
veterans through datasets and fiscal years and accumulated at the individual veteran level
since each veteran has a unique identifier (Gellad et al., 2018). The system uses the Patient
Treatment File that contains both acute and nonacute hospitalisation data on the patients
discharged from facilities that use VA software. Several hospitals across the United States
currently supply this database. This file is further divided into smaller subdivisions to give
specific information on the various sections of veterans' treatment process. The subdivisions
are principal, bed section, procedures, and surgery. These subdivisions contain different
pieces of information that help in processing the whole system.
The central subsection has information based on demographic data such as sex and
age of the patient, diagnostic and summary of the information in the related stages of
medication (Kohli & Tan 2016). The Bedsection file includes information on which treatment
the patient received and for how long. Based on this kind of information, a veteran can have

INDICATORS OF PATIENT SAFETY 5
more than one section record in one hospitalisation. The Procedure subfile contains data
relating to all the procedures that the patient underwent during his or her stay in the facility.
Such information includes date, time, and place where the procedures were done (Mages &
Kubic 2016). Finally, the Surgery subfile has data on every surgical procedure the patient
went through.
Modification of VA Software
Data required for PSI software
The VA software uses algorithms that link various pieces of data from different
subsections of the patient's treatment process to come up with the PSI event rates. For
instance, the software brings together codes on procedure and diagnosis with other pieces of
information different standard data sets (Peterson et al., 2018). This software requires some
raw data to generate the intended rates, such data may include but not limited to hospital
identification number, age, date, admission type length of stay, and secondary diagnosis
codes. It should be noted that the software uses data presented in a statistical format such as
Statistical Analytical System or Statistical Package for the Social Science.
Modifications to VA data
Most of the data types like age, sex hospital identifier, race as well as primary and
secondary diagnosis used in the software are already in the VA files and therefore required no
modification. However, there are some types of data that such as principle procedure,
admission type and days from admission to procedure need to be modified in the same format
as the other applicable data (Wilner & Ghassan, 2017). The illustrations below show how the
modification of each of the above data can be achieved. Additionally, some pieces of data

INDICATORS OF PATIENT SAFETY 6
were initially missing in the software and had to be incorporated just like the existing data
elements; like the admission source.
Comparison between VA and other software
The four files used in the VA data system have much more information about a
patient as compared to the files that are widely used Healthcare Cost and Utilization Project
(HCUP) data. Some of the areas that make VA software more superior include its ability to
link many pieces of information about a patient, like the various hospitalisations the patient
had. Another area making VA more suitable is the fact that it contains data from bed section
which give information on discharge time, procedures types and times, admission dates and
the primary and secondary diagnosis codes. As illustrated by Sutton, Eborall, and Martin
(2015), VA software accommodates both acute and noacute care while the Healthcare Cost
and Utilization Project contains only acute care.
The main aim of this study was to formulate test the methods of putting PSIs into use
through VA discharge data and at the same time acknowledging the differences and
similarities between rates of Patient Safety Indicators with VA and those with no VA
discharge data. The various modifications on VA inpatient administrative database were
meant to reduce the differences between VA database and the other software that perform the
same function but do not use Veteran Affairs databases. It should be noted that the
modifications solved the first challenge that some of the basic requirements were missing or
given a different definition from the ones that could be read by the software (Wilner, &
Ghassan 2017). It was also clear that the various modifications had different impacts on the
database characteristics and the PSI rates. For example, alterations like designating principle
procedure and admission impacted more than other changes. It is worth noting that the
various rules used in designing principle procedure and all other secondary procedures ended

INDICATORS OF PATIENT SAFETY 7
up affecting both the numerators and denominators of the final PSI rates. At the same time,
the algorithms used affected both elective admission rates as well as the other PSI rates.
We realised that the leading cause of the difference between the VA databases and
non-VA databases is the fact that the VA database has nonacute care while non-VA databases
have only acute care. For this reason, the exclusion of the nonacute care was necessary to
create a level ground for comparing the two sets of databases then reaggregate the remaining
acute parts of the mixed hospitalisations (Xie, Li, Swartz, & DePriest 2014). The new
features on the reaggregated admissions resulting from the alterations made on file structures
further required for more changes in the affected data elements. Such changes included
principle procedure, length of stay and principal diagnosis. This moderation between VA and
other sets of databases resulted in a levelled comparison ground even though it reduced the
amount of information in the VA database and diluted its actual image.
The fact that there were evident changes in the PSI rates between the VA database
with acute care and that with both acute and nonacute care dhow that the rates significantly
depend on the data. Another conclusion drawn from this scenario was the ability of the VA
database to identify potential patient safety with or without acute care setting (Reimer,
Schiltz, & Madigan 2016). The variations in the PSI rates between the original VA database
and the acute only VA database may be attributed to the changes in principle procedures,
principal diagnosis, and some procedures codes used. Any change in any of the elements of
hospitalisation record leads to either inclusion or exclusion in the PSI numerator or
denominators.
Challenges
From the above illustrations, the application of Veteran Affairs software in measuring
patient safety in a healthcare centre has more advantages that the application of other

INDICATORS OF PATIENT SAFETY 8
methods according to O’Brien et al. (2015). However, many critiques have argued that the
method carries risks of capturing wrong data from false events. The modifications applied to
the VA database may not work in other systems that try to adjust to implement the PSI
software. Our results suggest that other healthcare systems may face similar needs to make
modifications to their data to apply the PSI software (Qiu & Wang 2018). Even though
previous man studies indicate that there is the possibility of modifying the element data to
achieve compatibility in other systems in the healthcare industry, the resulting changes may
affect the comparison between the final PSI event rates. Most of the studies considered here
did not establish the risk-adjustment factors to derive how flexible the system works. This in
return paves the way for further studies in future in that area of research.
PDSA cycle
According to Hill, Lind, and Daraiseh (2015), the use of VA software as a patient’s
safety indicator in the hospital setup implies some level of improvement in performance of
the healthcare facility. This development fits into the PDSA cycle as illustrated bellow.
a. Planning Stage
At this stage, thorough study is done to learn the VA discharge database software in
details, its applicability, and the impacts the software would have on the regular operations of
the hospital (Kohli & Tan 2016). The decision to use the software majorly depends on the
goal being achieved, improving the patient safety in a hospital.
b. Do or testing stage

INDICATORS OF PATIENT SAFETY 9
This stage involves applying the actual program or VA discharge database software in
this case (Wilner & Ghassan 2017). The software is installed and allowed to run within the
already existing structure in the hospital.
c. Study
The first stage of the actual application should be studied over some time to analyse
the real impacts the software has in the hospital (Qiu & Wang 2018). From the illustrations
above, the VA software was confirmed to be working well with the other systems in the
hospital
d. Action time.
With no much alterations needed based on the analysis stage, the Veteran Affairs
discharge database software can be fully implemented in the healthcare centre to improve the
safety patients and other users of the facility (Kohli & Tan 2016).
Conclusion
From the above study, we realise that patients are at risk of experiencing adverse
events when exposed to some systems in healthcare centres. Such events can either be either
provider level indicators or area level indicators. One of the effective ways of measuring the
Patient Safety Indicators is the use of Veteran Affairs software that uses data from Patient
Treatment Files which categorises data into for subheadings namely main, bed section,
procedures, and surgery. The VA database software is considered advantageous in that it
accommodates both acute and nonacute care hospitalisations. However, the application raises
some concerns in that we have demonstrated the sensitivity of PSI rates to differences in data
file structure and definitions and sources of data elements (Muhammad Zia Hydari, Telang,

INDICATORS OF PATIENT SAFETY 10
& Marella 2015). The consequences of this sensitivity are amplified by the fact that PSI rates
are inherently low: most PSI rates are in the range of one to five per thousand
hospitalisations. Therefore, differences in data structures and algorithms that add or subtract
just one or a few cases from the numerator of a PSI for a given population and period could
make a meaningful difference in the overall PSI rate.

INDICATORS OF PATIENT SAFETY 11
References
Alyousef, B., Carayon, P., Hoonakker, P., Hundt, A. S., Salek, D., & Tomcavage, J. (2017).
Obstacles Experienced by Care Managers in Managing Information for the Care of
Chronically Ill Patients. International Journal of Human-Computer Interaction, 33(4),
313–321. https://doi.org/10.1080/10447318.2016.1270017
Carls, G. S., Henke, R. M., Karaca, Z., Marder, W. D., & Wong, H. S. (2015). The
Relationship between Local Economic Conditions and Acute Myocardial Infarction
Hospital Utilization by Adults and Seniors in the United States, 1995-2011. Health
Services Research, 50(5), 1688–1709. https://doi.org/10.1111/1475-6773.12298
Clark, R. E., Weinreb, L., Flahive, J. M., & Seifert, R. W. (2018). Health Care Utilization and
Expenditures of Homeless Family Members Before and After Emergency
Housing. American Journal of Public Health, 108(6), 808–814.
https://doi.org/10.2105/AJPH.2018.304370
Cohen, S. B., Cohen, J. W., Stagnitti, M. N., & Lefkowitz, D. C. (2016). Implementation of a
linked Medical Organization Survey in the Medical Expenditure Panel
Survey. Journal of Economic & Social Measurement, 41(4), 417–432.
https://doi.org/10.3233/JEM-170436
Dorflinger, L. M., Ruser, C., Sellinger, J., Edens, E. L., Kerns, R. D., & Becker, W. C.
(2014). Integrating Interdisciplinary Pain Management into Primary Care:
Development and Implementation of a Novel Clinical Program. Pain
Medicine, 15(12), 2046–2054. Retrieved from
http://search.ebscohost.com/login.aspx?
direct=true&db=s3h&AN=100100908&site=ehost-live

INDICATORS OF PATIENT SAFETY 12
Helm, J. E., Alaeddini, A., Stauffer, J. M., Bretthauer, K. M., & Skolarus, T. A. (2016).
Reducing Hospital Readmissions by Integrating Empirical Prediction with Resource
Optimization. Production & Operations Management, 25(2), 233–257.
https://doi.org/10.1111/poms.12377
Hill, A. K., Lind, M. A., Tucker, D., Nelly, P., & Daraiseh, N. (2015). Measurable results:
Reducing staff injuries on a specialty psychiatric unit for patients with developmental
disabilities. Work, 51(1), 99–111. https://doi.org/10.3233/WOR-152014
Gellad, W. F., Thorpe, J. M., Xinhua Zhao, Thorpe, C. T., Sileanu, F. E., Cashy, J. P., …
Fine, M. J. (2018). Impact of Dual Use of Department of Veterans Affairs and
Medicare Part D Drug Benefits on Potentially Unsafe Opioid Use. American Journal
of Public Health, 108(2), 248–255. Retrieved from
http://search.ebscohost.com/login.aspx?
direct=true&db=s3h&AN=127233472&site=ehost-live
Kohli, R., & Tan, S. S.-L. (2016). Electronic Health Records: How Can Is Researchers
Contribute to Transforming Healthcare? MIS Quarterly, 40(3), 553–574. Retrieved
from http://search.ebscohost.com/login.aspx?
direct=true&db=buh&AN=117951008&site=ehost-live
Mages, R., & Kubic, T. T. (2016). Counterfeit medicines: Threat to patient health and
safety. Pharmaceuticals Policy & Law, 18(1–4), 163–177.
https://doi.org/10.3233/PPL-160441
Muhammad Zia Hydari, Telang, R., & Marella, W. M. (2015). Electronic Health Records and
Patient Safety. Communications of the ACM, 58(11), 30–32.
https://doi.org/10.1145/2822515

INDICATORS OF PATIENT SAFETY 13
O’Brien, W. J., Chen, Q., Mull, H. J., Shwartz, M., Borzecki, A. M., Hanchate, A., & Rosen,
A. K. (2015). What Is the Value of Adding Medicare Data in Estimating VA Hospital
Readmission Rates? Health Services Research, 50(1), 40–57.
https://doi.org/10.1111/1475-6773.12207
Peterson, K., Anderson, J., Boundy, E., Ferguson, L., McCleery, E., & Waldrip, K. (2018).
Mortality Disparities in Racial/Ethnic Minority Groups in the Veterans Health
Administration: An Evidence Review and Map. American Journal of Public
Health, 108(3), e1–e11. Retrieved from http://search.ebscohost.com/login.aspx?
direct=true&db=s3h&AN=127874093&site=ehost-live
Qiu, B., & Wang, T. (2018). Does Knowledge Protection Benefit Shareholders? Evidence
from Stock Market Reaction and Firm Investment in Knowledge Assets. Journal of
Financial & Quantitative Analysis, 53(3), 1341–1370.
https://doi.org/10.1017/S0022109018000066
REIMER, A. P., SCHILTZ, N., KOROUKIAN, S. M., & MADIGAN, E. A. (2016). National
Incidence of Medical Transfer: Patient Characteristics and Regional
Variation. Journal of Health & Human Services Administration, 38(4), 509–528.
Retrieved from http://search.ebscohost.com/login.aspx?
direct=true&db=buh&AN=113300581&site=ehost-live
Rosen, Amy K., Qi Chen, Ann M. Borzecki, Marlena Shin, Kamal M. F. Itani, and Michael
Shwartz. 2014. “Using Estimated True Safety Event Rates versus Flagged Safety
Event Rates: Does It Change Hospital Profiling and Payment?” Health Services
Research 49 (5): 1426–45. doi:10.1111/1475-6773.12180.

INDICATORS OF PATIENT SAFETY 14
Spitzer, S. A., Staudenmayer, K. L., Tennakoon, L., Spain, D. A., & Weiser, T. G. (2017).
Costs and Financial Burden of Initial Hospitalizations for Firearm Injuries in the
United States, 2006-2014. American Journal of Public Health, 107(5), 770–774.
https://doi.org/10.2105/AJPH.2017.303684
STAVROPOULOU, C., DOHERTY, C., & TOSEY, P. (2015). How Effective Are Incident-
Reporting Systems for Improving Patient Safety? A Systematic Literature
Review. Milbank Quarterly, 93(4), 826–866. https://doi.org/10.1111/1468-
0009.12166
Sutton, E., Eborall, H., & Martin, G. (2015). Patient Involvement in Patient Safety: Current
experiences, insights from the wider literature, promising opportunities? Public
Management Review, 17(1), 72–89. https://doi.org/10.1080/14719037.2014.881538
Wilner, S. J. S., & Ghassan, A. (2017). Tales of seduction and intrigue: design as narrative
agent of brand revitalisation. Journal of Marketing Management, 33(3/4), 173–202.
https://doi.org/10.1080/0267257X.2016.1271346
Xie, X., Li, J., Swartz, C. H., & DePriest, P. (2014). Improving Response-Time Performance
in Acute Care Delivery: A Systems Approach. IEEE Transactions on Automation
Science & Engineering, 11(4), 1240–1249.
https://doi.org/10.1109/TASE.2013.2258913