HOSPITAL ACQUIRED CONDITIONS RESEARCH 2

Hospital-Acquired Condition Research

Hospital-acquired condition (HAC) is a condition or an undesirable situation that affects
a patient and arises during their stay in a medical facility (Watson Health, 2018). Centre for
Medicare and Medicaid Services (CMS) defines HACs as avoidable care complications that,
with the application of guidelines that are evidence-based, can reasonably be prevented (Watson
Health, 2018). Even though infections form the bulk of hospital acquired conditions, there are
many other conditions and situations that are considered as hospital-acquired conditions. They
include falls and trauma, mismatch of blood type, and pressure ulcers (Watson Health, 2018).
Since 2008, HACs have been used by CMS to determine the amount of reimbursements that
hospitals receive. The more HACs a hospital has the less the reimbursement they receive. This
approach is meant to encourage hospitals to take maximum care of patients to prevent them from
getting affected by HACs. It is assumed that the perceived avoidability of HACs and CMS
penalties using reimbursements, hospitals and medical care facilities have enough reasons to
reduce HACs. However, research shows that HACs remain high in hospitals and the associated
costs to patients also remain high. The findings suggest that current financial incentives are
perhaps not effective in motivating hospitals to reduce HACs.
Purpose of the research
The research titled “Research brief: hospital acquired conditions lead to avoidable cost
and excess deaths” was conducted by IBM Watson Health. It sought to determine the prevalence
of HACs in hospitals, their costs, and, therefore, the effectiveness of Hospital Acquired
Condition Reduction Program (HACRP) that is run by Medicare. The program contains 14
HACs which monitors. Each year, it analyzes HAC statistics of participating hospitals to

HOSPITAL ACQUIRED CONDITIONS RESEARCH 3
determine how they perform. Hospitals that are among the worst performing 25% have their
hospital reimbursement reduced by 1% (Watson Health, 2018).
Source of data for the research
In conducting the research, IBM Watson Health used data drawn from its Projected
Inpatient Database of 2016. The database has data from 2,600 acute care hospitals. From these
hospitals it has data of around 20 million discharges that are projected statistically to the whole
of the United States (Watson Health, 2018). The data set is very comprehensive as it contains
data of claims made by all payers countrywide. These payers include health plans that are
employer sponsored as well as private payers and CMS. In this particular research, IBM Watson
Health used 19 million actual hospital discharges and projected them to 37 million discharges
from around 4,500 medical facilities and hospitals across the country (Watson Health, 2018).
Types of descriptive statistics and graphical representations of data that were used
Descriptive statistics involve use of brief descriptive coefficients to provide a summary of
a given set of data (Holcomb, 2016). The summarized data can be either of an entire population
or just a sample. There are four main types of descriptive statistics. They are measures of
frequency, measures of central tendency, measures of position, and measures of variation or
dispersion (Holcomb, 2016). Measures of frequency shows how frequently something occurs. It
includes statistics such as frequency and percentage. Measures of central tendency shows a
response or figure that is most commonly indicated. It includes statistics such as mode and mean.
Measures of position include statistics such as quartile and percentile ranks. They show how
given scores relate to other scores in a given set of data. The last type is measures of dispersion

HOSPITAL ACQUIRED CONDITIONS RESEARCH 4
or variation. These measures include standard deviation, range, and variance. They identify
scores’ spread by stating the intervals of the scores (Holcomb, 2016).
In the research reviewed, only two descriptive statistics were used. These are measures of
frequency and measures of central tendency. In the former, the research used percentages. It was
found that hospital acquired conditions (HACs) increased the death rate of a patient by 72%. The
measure of central tendency statistic that was used is mean. The research found that HACs
increased the number of days that patients spent in hospital by an average of 8 days (Watson
Health, 2018).
In addition to these descriptive statistics, the research also used graphs to represent data.
Graphical representation of data makes the data easy to understand. It also make comparison
with other data simpler. In the research reviewed, the authors used graphs to show the number of
cases of hospital acquired conditions in 2016. They also used them to show the number of
avoidable deaths caused by HAC.
What were the research question(s) and the significant findings of the article?
The research did not have a question or questions. However, it sought to find out the
prevalence and impact of hospital acquired conditions (HACs) to both patients and hospitals.
Despite efforts by CMS and hospitals to reduce cases of HACs, the research found that
HAC cases remain high in the country. In 2016, a total of 48,771 cases were reported in the USA
(Watson Health, 2018).
It also found that HACs had significant financial and health impact on patients and
hospitals. HACs resulting in an increase of hospital care of around $2 billion. This figure

HOSPITAL ACQUIRED CONDITIONS RESEARCH 5
translates to $41,000 per patient who was affected by hospital acquired condition. The costs to
hospitals are likely to be higher because the $2 billion figure does not include Medicare’s
penalties under the HACRP program (Watson Health, 2018).
The health impact of HACs was also dire. It was found that 3,219 deaths that could be
avoided occurred as a result of HACs. HACs increased patient mortality rate by 72%. In
addition, they led to patients spending more time in hospitals. On average, patients with HACs
spent an extra 8 days in hospital (Watson Health, 2018).
How the information presented could be used to inform decisions or improvements
The findings of the research are valuable both for hospitals and Medicare. When
Medicare introduced the HACRP program, it was sold as the program that would significantly
reduce the number of HAC cases in hospitals. The findings show that perhaps the method
employed to encourage this reduction has not been effective thus far as HAC cases remain quite
high.
To encourage hospitals to reduce HAC cases, Medicare’s HACRP program penalizes
hospitals that record poor HAC performance in a given year. Regardless of how well a hospital
has improved as long as its HAC performance is among the last quarter of hospitals, Medicare
penalizes it.
It is not clear from the findings whether this method is the cause of poor performance of
HACRP in reducing HAC cases in hospitals. However, what is clear is that the pay-per-
performance method employed by Medicare is not providing a strong enough incentive for
hospitals to adopt measures that reduce HAC cases. Thus, information from this research could
be used by Medicare to better structure its HACRP programme so that it becomes better at

HOSPITAL ACQUIRED CONDITIONS RESEARCH 6
encouraging hospitals to reduce HAC cases. Perhaps Medicare should think of replacing its
value-based payment model with another method that has proved to be successful elsewhere.
Hospitals will also find the findings useful. The high extra costs and many unnecessary
deaths resulting from HACs should jolt them into putting in place measures that reduce HAC
cases. The findings show that whatever measures that they are currently using are not effective.

HOSPITAL ACQUIRED CONDITIONS RESEARCH 7

References

Health, W. (2018). Research Brief: Hospital Acquired Condition lead to avoidable costs and
excess deaths. IBM Watson Health.
https://googleweblight.com/i?u=https://www.ibm.com/blogs/watson-health/research-
brief-hospital-acquired-conditions-lead-to-avoidable-cost-and-excess-deaths/&hl=en-
KE&tg=98&pt=3
Holcomb, Z. C. (2016). Fundamentals of descriptive statistics. Routledge.