Evidence-Based Practice: Nursing Workload and Adverse Events Review

This assignment critically appraises a systematic review by Oliveira, Garcia, and Nogueira (2016) which investigates the impact of nursing workload on the occurrence of adverse events in intensive care units (ICU). The report employs a PICO strategy (Patient, Intervention, Comparison, Outcome) to analyze the research question, focusing on adult ICU patients, nursing workload as an intervention, and adverse events as outcomes. The paper reviews the methodology, including the use of various databases like MEDLINE, CINAHL, and LILACS, and the application of STROBE guidelines to ensure the quality of included articles. It assesses the rigor of the review, data synthesis methods, and the combination and display of results. The findings highlight the association between heavy nursing workloads and increased adverse events, underscoring the importance of adequate staffing and workload management in ICU settings to ensure patient safety and improve healthcare outcomes. The review synthesizes findings from multiple studies, discussing variations in results and the causal factors behind them, offering insights into the complexities of workload and its impact on patient care.

Running head: Evidence-Based Practice 1
Evidence-based Practice
by
Course:
Tutor:
University:
Department:
Date:

Evidence-Based Practice 2
Nursing workload and occurrence of adverse events in intensive care: a systematic review
1.0 Introduction
The heavy workload for healthcare nurses is a global health problem that is caused by multiple
factors. Some of the common factors that lead to heavy nursing workload in intensive care unit
(ICU) include high demand for nurses, increased overtime, reduced staffing, limited supply of
nurses and decrease in the length of stay of patients among others. Nursing workload has several
adverse effects on patients in the intensive care unit. The ICU is a department meant for the care
of critically sick patients, and there are high chances of the occurrence of adverse events.
Nursing workload has been defined as “work process elements that interact dynamically with
each other and with the body of the worker, generating a process of adaptation which leads to
wear” (Daud-Gallotti et al., 2012). Patient safety and nursing care in the ICU is dependent on the
qualification of the experts and an adequate supply of human resources. Any mismatch in the
number of healthcare experts and the recommended care by the patients in ICU can result to
work overload and failures in patient safety (Liu et al., 2012). Therefore, an evaluation of
nursing workload in ICU is critical for patient care and safety. The objective of this paper is to
critically review a systematic review by Oliveira, Garcia, and Nogueira (2016) on the effect of
nursing workload on the frequency of adverse effects in ICU patients.
2.0 Review Problem
A PICO (patient, intervention, comparison and outcomes) strategy was used to analyze the
research question.
Population
The study considered adult patients admitted to the intensive care unit.

Evidence-Based Practice 3
Intervention
The nursing workload was used as an intervention.
Outcomes
The considered outcomes were adverse effects (patient safety).
3.0 Literature Review
3.1 Appropriateness of the Articles
The articles included in the review were relevant to the research question considering the
inclusion and exclusion criteria and the use of the Cochrane study selection procedure (Stovold,
Beecher, Foxlee, & Noel-Storr, 2014). Out of the identified 596 potential studies, only 8 of them
were included in the review. Furthermore, the use of STROBE guidelines in determining the
articles to be included or excluded is an indication that the articles included in the review were of
high quality (Von Elm et al., 2014). All the eight articles were appropriate for the research
question because all their objectives addressed the impact or association between nursing
workload and patient safety and care. Moreover, the articles considered ICU patients or severe
health conditions that are only handled at the ICU department. This ensured that the only
relevant and high-quality literature were included in the systematic review.
3.2 Study Design
The study used a systematic review of relevant literature. This design was appropriate for the
research question which was aimed at assessing an association between two observable
parameters: nursing workload and adverse effects. The design was appropriate because it allows
adherence to rigour, replicability and transparency which improves the strength of traditional
literature review in multiple ways. Systematic reviews are important in reducing implicit
researcher bias which is common in reviews due to the frequent citation of literature because

Evidence-Based Practice 4
they are often restricted to the literature already known by researchers. Researcher bias is
reduced through the use of broad search strategies, use of inclusion and exclusion criteria and
predefined search strategy thus forcing the researchers to search for literature outside their areas
of specialization and networks (Mallett, Hagen-Zanker, Slater, & Duvendack, 2012).
3.3 Usage of Appropriate Database and Outside Resources
The study used multiple databases to collect data through an electronic search. The Medical
Literature Analysis and Retrieval System Online (MEDLINE) was used through PubMed to
search for relevant articles. MEDLINE is a bibliographic database of life sciences and contains
academic journals in the healthcare sector. The appropriate key terms were used and combined
using Boolean operators “OR” and “AND”. The largest number of studies (n=226) were
retrieved from MEDLINE. The descriptors were obtained from the Medical Subject Headings
(MESH). The Cumulative Index to Nursing and Allied Health Literature (CINAHL) was another
database that was used in the review. CINAHL is an index of journal articles published in the
English language and are for nursing and healthcare. Boolean operators were used to combining
the keywords and the database generated 226 articles making it the second-largest source of
articles.
The Latin American and Caribbean Center on Health Sciences Information (LILACS) also had
the third-largest number of studies (n=14) and was published in Portuguese and the keywords
combine using Boolean operators. Other databases used include the database of nursing
(BDENF) (n=6); and the Cochrane library (n=2). The portal Scientific Electronic Library Online
(SciELO) was analyzed as well and 3 articles retrieved. Furthermore, other databases were
examined such as Google Scholar alongside a manual analysis of the bibliography in the selected
articles. The use of diverse and several databases ensured that the best quality articles were

Evidence-Based Practice 5
included in the study and hence improving the reliability of the findings (Leung, 2015).
Furthermore, the selected databases were appropriate because they only contained journals about
healthcare, allied health, biomedicine and nursing.
4.0 Quality of the Review
4.1 Rigor
There is multiple evidence of the authors’ efforts to assess the quality of the studies. The
selected articles were assessed by examining the title, author, journal, the year and place of
publication, objective, research design, a measurement tool for workload, general findings and
conclusions. Furthermore, there was an independent application of the instrument strengthening
the reporting of observations studies in Epidemiology (STROBE) by two researchers. An
independent review provides an opportunity for more accuracy and criticism on the articles, thus
increasing their reliability. Moreover, an independent review by the two researchers ensures
confirmability which also improves the validity of the study. Triangulation method has also been
used in the study to achieve rigour. The researchers used over five different databases to source
data. This approach improves the quality of the findings because it aims at overcoming intrinsic
bias often caused by using a single method (Ghafouri, & Ofoghi, 2016).
The quality of the studies was also improved by using the inclusion and exclusion criteria. The
included articles must have been fully published in English, Spanish or Portuguese and in
international and national journals. Moreover, they must have utilized the measurement tools for
nursing workload in adult patients in ICU. All other articles, not peer-reviewed and not fully
published were excluded from the study. This ensured that only expert journal articles were
analyzed, thus increasing the credibility and reliability of the outcomes (Leung, 2015).

Evidence-Based Practice 6
4.2 Data Synthesis
The abstract and the titles of the studies obtained from the search strategy were analyzed to
determine those relevant to the topic of study. The research question of those articles with
insufficient data was analyzed to determine whether they were to be included and to avoid
erroneous exclusions. Two researchers analyzed the preselected studies in their entirety after
recovery. However, if both of them disagreed, then a third researcher was brought on board to
analyze the research and to determine the dominant view of the majority. Ritchie, Lewis,
Nicholls, and Ormston (2013) observes that an independent review in research ensures that there
is accuracy in the findings and thus making the outcomes more credible and reliable. Data
extraction from the studies was undertaken using a tool with the following sections: the title,
author, the year of publication and place of the journal, the objectives of the study, research
design, the measurement instrument of nursing workload, nature of the adverse event, primary
outcomes and conclusions. The use of a standardized tool to extract data ensures that there is
consistency in the analysis of the studies since they are all subjected to a standard procedure.
This enables easy comparison of the studies and improves the reliability of the findings (Morse,
2015).
The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE)
instrument was used by two independent researchers to examine the quality of the included
articles. The tool was used as a guide to determine the components of a more accurate and
complete observational article. STROBE analysis is significant in the description of sample
studies (Langan et al., 2013).
4.3 Combination of the Results
The outcomes of the studies were all combined under different themes and sub-themes. 8 out of
37 selected studies were included in the analysis. 75.0% of the included articles were obtained

Evidence-Based Practice 7
from international journals whereas 25.0% were obtained from national journals. 75.0% of the
articles were prospective studies whereas 25.0% were retrospective investigations. The
frequency of women included in the studies ranged from 50.9 to 65.6 years. Under the type of
instrument workload used, the Nursing Activities Score and the Therapeutic Intervention Scoring
System were used by 37.5% of the articles with 25.0% of the studies using the Omega Scoring
System and PRN. Other combined findings included the types of adverse events. The
combination of the outcomes makes it easy to analyze the outcomes of each study and to
determine the overall findings and any research gap in the topic of study (Flick, 2018).
4.4 Display of Individual Studies
The findings of each study were displayed in the form of a table under different areas such as the
characteristics of the sample patients, adverse events evaluated, the instrument of nursing
workload used, findings, and the impact of nursing workload on the frequency of adverse events.
The display of each study in the form of a chart made comparison easy (Ulin, Robinson, &
Tolley,2012).
4.5 Similarities among Individual Studies
The review did not include any similarities among the individual studies
4.6 Consideration of Similar Results from Different studies
The review considered similar findings across the eight articles included for analysis. This was
achieved by combining the outcomes of the studies. For instance, 75.0% of the articles were
prospective studies whereas 25.0% were retrospective investigations, the Nursing Activities
Score and the Therapeutic Intervention Scoring System were used by 37.5% of the articles with
25.0% of the studies using the Omega Scoring System and PRN among others. The consideration
of similar studies across the included articles helps to show areas of emphasis and helps to
distinctly answer the research question (Mabuza, Govender, Ogunbanjo, & Mash, 2014).

Evidence-Based Practice 8
4.7 Variations in the Results
The variations in the results were discussed alongside the causal factors. For instance, there were
discrepancies regarding the mean NAS score about international journals. An estimated 41% of
Spanish studies used NAS, but over 95% were used in Norwegian studies (Stafseth, Solms, &
Bredal, 2011). The causal factors for these variations in the national and international journals
included the clinical features of the patients, certain characteristics of the ICU, and the variation
in the understanding and use of the instrument.
5.0 Results
5.1 Overall Results (numerical)
75.0% of the articles were retrieved from international journals and 25.0% from national
journals. The cohort prospective studies were the majority (75.0%) and only 25.0% were
retrospective. The studies had a sample of 79 to 970 patients in the ICU with women being the
majority and aged between 50.9 to 65.6 years. 37.5% of the studies used the NAS, TISS (37.5%)
and 25.0% used Omega and PRN. Different types of adverse effects were also observed in the
studies. 62.5% of the studies used healthcare-associated infections, medication errors were
examined in 25.0% of the studies and 25.0% of the studies used the occurrence of pressure ulcer.
The Chilean study used sentinel incidents which consisted of AE such as accidental loss of
central venous catheter, falls etc (Oliveira et al., 2016). A nursing workload means a range of
61.97% to 81.2% was observed in the Brazil studies that used NAS. On the other hand, over 95%
of the studies were carried out in Norway and 41% in Spain (Stafseth et al., 2011).
5.2 Precision of the Results
Since the study was a systematic review, its precision was greatly dependent on the articles
included in the analysis. The reviewed articles had a high precision of the results as evidenced by

Evidence-Based Practice 9
the use of odds ratio, p-values and confidence intervals. An Odds ratio measures the relationship
between an exposure and an outcome and is effective in determining whether a given exposure is
a risk factor for an outcome or not (Grant, 2014). On the other hand, a confidence interval is used
to determine the precision of the odds ratio. However, the 95% CI used doesn’t report a
statistical significance of any measure (Cumming, 2013). The p-values were also used to
determine the statistical significance of the variables under study. The analysis of articles which
have used p-values, odds ratio and CI ensures that the findings of the analysis have high
precision (Greenland et al., 2016).
5.3 Expression of the Results
The outcomes have been presented using percentages and corresponding numerical values. This
makes comparison easy and provides quantitative evidence of the outcomes. However, the
percentages and numerical figures do not accurately present the causal factors behind the
discrepancy and therefore are ineffective in comparison and drawing of meaningful results (Lee,
2016).
6.0 Discussion
6.1 Application to the Local Population
The outcome of the studies applies to the local population because of multiple reasons. The study
has detailed methodology on how the review was conducted thus making it possible to be
replicated or transferred to in any local population with relatively similar characteristics (Smith,
2018). Although the participants consisted of adult patients in ICU, it doesn’t limit its application
to children patients in ICU because nursing workload in the intensive care environment is
relatively the same.

Evidence-Based Practice
10
6.2 Consideration of all-important outcomes
The review considered all significant results. This can be deduced from the research question
which was fully answered. The objective of the study was to find out whether the nursing
workload influenced the incidence of adverse effects in adult patients admitted to the ICU. The
study found out that nursing workload impacted the incidence of AE in adult patients in the ICU.
Additionally, the themes obtained from the analysis are an indication of the complete
presentation of the outcomes. For instance, the study outlined the instruments used to measure
nursing workload, factors that cause the discrepancy in the levels of nursing workload, types of
adverse effects (Oliveira et al., 2016).
6.3 Advantages and Disadvantages of the Results
The outcomes of the study are beneficial to the healthcare setting in multiple ways. The study
shows the existence of an association between nursing workload and adverse effects in ICU
patients. The results indicate that addressing nursing workload is one effective way of improving
patient safety and care through the prevention of injury and other risk factors such as severity,
length of stay in the ICU, higher welfare costs, and high mortality rates. The findings are also
critical because they demonstrate the need for increasing the number of nursing professionals as
a strategy of achieving quality care. However, the findings cannot wholly be relied upon because
there is the need for larger samples with extensive follow-up period to strengthen the evidence of
the association between the nursing workload and various adverse effects in ICU (Oliveira et al.,
2016).

Evidence-Based Practice
11
6.4 Recommendations for Future Research
There is a need for further research on multicenter studies with larger samples and extensive
follow-up to provide more evidence on the association between nursing workload and the
incidence of different forms of adverse effects in the intensive care unit.

Evidence-Based Practice
12
References
Cumming, G. (2013). Understanding the new statistics: Effect sizes, confidence intervals, and
meta-analysis (11th ed.). New York, NY: Routledge.
Daud-Gallotti, R. M., Costa, S. F., Guimarães, T., Padilha, K. G., Inoue, E. N., Vasconcelos, T.
N., ... & Levin, A. S. (2012). Nursing workload as a risk factor for healthcare associated
infections in ICU: a prospective study. PloS one, 7(12), e52342.
Flick, U. (2018). An introduction to qualitative research (4th ed.). California, CA: Sage
Publications Limited.
Ghafouri, R., & Ofoghi, S. (2016). Trustworth and rigor in qualitative research. International
Journal of Advanced Biotechnology of Applied Behavioral Science, 7, 90-101.
Grant, R. L. (2014). Converting an odds ratio to a range of plausible relative risks for better
communication of research findings. Bmj, 348, f7450.
Greenland, S., Senn, S. J., Rothman, K. J., Carlin, J. B., Poole, C., Goodman, S. N., & Altman,
D. G. (2016). Statistical tests, P values, confidence intervals, and power: a guide to
misinterpretations. European journal of epidemiology, 31(4), 337-350.
Langan, S. M., Benchimol, E. I., Guttmann, A., Moher, D., Petersen, I., Smeeth, L., ... & Von
Elm, E. (2013). Setting the RECORD straight: developing a guideline for the REporting
of studies Conducted using Observational Routinely collected Data. Clinical
epidemiology, 5, 29.