Describe the data and the results of any statistical tests or analyses presented in the article
NOTE: We do not resell pre-written papers. Upon ordering a paper, we custom-write an original paper exclusively for you. Please proceed and order an original paper to enjoy top grades.
Order a Similar Paper
Order a Different Paper
Initial posting needed in 14 hours Minimum of 350 words
Post an APA citation and brief summary of the research article that you selected. Selected article is
Humphreys, H., Becker, K., Dohmen, P. M., Petrosillo, N., Spencer, M., van Rijen, M., … & Garau, J. (2016). Staphylococcus aureus and surgical site infections: benefits of screening and decolonization before surgery. Journal of Hospital Infection, 94(3), 295-304.Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/27424948
1. Describe the data and the results of any statistical tests or analyses presented in the article.
2. Explain how the researchers formulated their conclusion, any weaknesses in their analysis or conclusions, and offer at least one alternate interpretation of their data.
3. Propose at least one additional research study that could be done to further investigate this research topic.
Read sections of the chosen article where the data is presented, analyzed, and interpreted for meaning.
- What reasoning process did the researchers use to formulate their conclusions? What explanation did they give to support their conclusions? Were there any weaknesses in their analysis or conclusions?
- Consider possible alternate conclusions that the researchers could have drawn based on their data.
- Examine the findings that the article presents and consider how well they addressed the researcher’s initial question(s). What additional research could be done to build on these findings and gain a fuller understanding of the question?
Read a selection of your colleagues’ responses.
Responses needed in 36 hours Minimum of 150 words
Respond to at least two colleagues on two different days using one or more of the following approaches:
- Choose a section of your colleagues post to respond to and expound on it.
- Ask a probing question, substantiate with additional background information, and evidence.
- Share an insight from having read your colleagues’ postings, synthesizing the information to provide new perspectives.
- Offer and support an alternative perspective using readings from the classroom or from your own review of the literature in the Walden Library.
- Expand on your colleagues’ postings by providing additional insights or contrasting perspectives based on readings and evidence.
Research is used by many disciplines to acquire new knowledge. However, the study must be designed correctly to yield valid results. As Rohrig, du Prel, and Blettner (2009) note, errors in study design cannot be changed after the study is completed. The purpose of this discussion post is to give a summary of the research done by Goldsack, Bergey, Mascioli, and Cunningham (2015), what statistical tests were used, how the statistics were analyzed, how conclusions by the authors were formed, weaknesses in the findings, and to offer one alternative interpretation of the results and one additional research study that can be done.
Summary of Article
Goldsack et al. (2015) did a 30-day prospective pilot study with pre- and post-implementation evaluation of hourly rounding effects on patient falls. The pilot period was September 23rd through October 20th of 2013 on two different medical units. Rounding was clarified to be checking on patients hourly between 0600 and 2200, then every two hours from 2200 to 0600. On Unit 1, both nurses and assistive personnel could round on patients, while Unit 2 only nurses could round on patients. Unit 1 also had a two-way teaching method and staff involvement in the program, while Unit 2 did not (Goldsack et al., 2015).
Data and Conclusions
Goldsack et al. (2015) measured falls as the number of falls per 1,000 patient days. In addition, compliance in rounding was measured by Goldsack et al. (2015) by managers randomly selecting flow sheets and reviewing recorded times of rounding for the previous 24 hours, gathering both the average and median times between rounds and calculated them, managers selected 60 rounds to observe at random, and the researchers used convenience sampling to survey one nurse from each unit for each shift on the round they just completed.
Statistical tests were used by Goldsack et al. (2015), such as the Mann-Whitney test to compare baseline fall rates and the project fall rates on Unit 1. Also, Goldsack et al. (2015) used the one-sample Wilcoxon-signed rank test to compare falls from project time to another time, such as Unit 1 baseline fall rate and the pilot fall rate, Unit 1 project fall rate and pilot fall rate, and Unit 2 project and pilot fall rates. Finally, Goldsack et al. (2015) used robust regression analysis to see whether the median intervals between hourly rounds was increasing, which would mean staff compliance was decreasing with hourly rounding. In addition, Goldsack et al. (2015) used robust regression analysis to determine if there was an increase in rounding percent with an increase in the percent of staff reporting they were rounding.
The conclusions Goldsack et al. (2015) found from their data analysis showed that Unit 1 had a decrease in falls from a baseline of 3.9 falls/1,000 patient days, to 1.3 falls/1,000 patient days during the pilot period, to 2.5 falls/1,000 patient days during the project period (p. 27). On Unit 2, Goldsack et al. (2015) found the baseline fall rate to be 2.6 falls/1,000 patient days and a fall rate of 2.5 falls/1,000 patient days during the pilot period (p. 27). In addition, Goldsack et al. (2015) found the mean time between rounds did not increase during the project period, which suggests there was no decrease in nurses doing hourly rounding. Goldsack et al. (2015) also sent out a survey for nurses to answer on hourly rounding for both units and found that 89% of nurses on Unit 1 would recommend hourly rounding on other groups versus 25% of nurses from Unit 2 (p. 28).
Goldsack et al. (2015) drew several conclusions based on the statistical analysis of baseline, pilot, and project fall rates, in addition to nurses responses on their survey. Overall, Goldsack et al. (2015) noted hourly rounding to decrease patient falls, however, other factors may have been at play, such as nurse involvement, which could account for Unit 1 having a very low fall rate during their pilot period. In addition, based on the survey results, Goldsack et al. (2015) found that leadership involvement and nurse involvement (such as a unit champion) in successful fall prevention implementation. However, Goldsack et al. (2015) note some weaknesses, such as a short pilot period, and a call for more research being needed to determine if hourly rounding itself can cause a decrease in patient falls, or if a different fall prevention program could work, as long as the staff is involved.
Alternatives to Data Interpretation
One alternative data interpretation of Goldsack et al. (2015) fall rates can include the reason for Unit 1’s fall rates being lowest during the pilot period could be because of staff’s awareness about hourly rounding and fall prevention, thus they are more likely to look for fall risks and be more proactive to prevent falls. Therefore, the increase in falls for Unit 1 between pilot and project periods could be because the staff is less aware of fall risks, and over time may have a gradual increase in fall rates.
Another alternative to data interpretation of Goldsack et al. (2015) for staff recommendations of hourly rounding to other units could be because of nursing assistant’s assistance with hourly rounding. Goldsack et al. (2015) let Unit 1 have either nurses or assistive personnel round, while Unit 2 only had nurses rounding, which could lead to nurses feeling unsatisfied with hourly rounding versus less leadership involvement.
An additional research study should be done to investigate further whether or not hourly rounding influences patient fall rates. I did not like how Goldsack et al. (2015) only did their research in one hospital, as I would like to see the study done in more hospitals at the same time. In addition, there was no statement of the sample size to know if the sample Goldsack et al. (2015) used was large enough to represent the population. Finally, as Laureate Education (2012) noted, Cronbach’s Alpha is a numerical coefficient used to reflect consistency and reliability, with a higher score meaning more reliability. Neither Cronbach’s Alpha or a sampling size was used in the research done by Goldsack et al. (2015), which can affect the reliability and validity of their results. Another research study should be done with a sampling size and Cronbach’s Alpha to make sure the research is consistent and reliable.
Research preparation takes time and effort to make sure the data and conclusions are meaningful and can have an impact on knowledge. Laureate Education (2012) noted that nurses need to know sound research methods and practices to evaluate research literature. Thus, nurses should be able to critique and question research before they decide to try and implement it into evidence-based practice.
Goldsack, J., Bergey, M., Mascioli, S., & Cunningham, J. (2015). Hourly rounding and patient falls: What factors boost success? Nursing2015, 25-30. doi: 10.1097/01.NURSE.0000459798.79840.95
Laureate Education. (Producer). (2012). Weighing the evidence. Baltimore, MD: Author
Rohrig, B., du Prel, J. -B. & Blettner, M. (2009). Study design in medical research: Part 2 of a series on the evaluation of scientific publications. Deutsches Aerzteblatt International, 106(11), 184-189. doi: 10.3238/arztebl.2009.0184
This purpose of this study was to show a possible relationship with increased nursing fatigue and work absences. The author referenced previous research proving fatigue and sleep deprivation leads to not only medical errors in the hospital, but also can jeopardize their health as well. The article highlighted causes of nursing fatigue which included: lack of sleep, loss of rest periods between shifts, longer shifts, added shifts, higher acuity levels, and working night shift. “Nurses fatigue scores for every worked shift were generated using the Fatigue Assessment Tool by InterDynaics (FAID) and Fatigue Risk Index (FRI) software programs.” (Sagherian et al, 2017).
FAID – Estimates fatigue-risk scores for every working shift based on the input of work schedules
– (2) steps: takes sleep-wake patterns from input of work-rest schedules, then generates fatigue
– Higher scores indicate higher fatigue exposures and less alertness
FRI – Cumulative total that incorporates: effect of previous shift, duty time of present shift, shift length,
and job type.
- Range between 0 – 100: Represent the average probability of high levels of sleepiness on the Karolinska Sleepiness Scale (KSS) which is a 9-point rating scale with 1 being extremely alert to 9 being extremely sleepy. (Ex. FRI score of 40, has 40% chance of scoring 8-9 on KSS)
Data and Results of Statistical Tests
Design: Retrospective Cohort Design
Population: 197 nurses working 12-hr shifts in mid-Atlantic Metropolitan pediatric hospital
Time period: 18 months – 43,893 shifts total
Descriptive statistics “describes and summarizes data”, were used to describe the characteristics of the study. (Polit & Beck, 2017).
Spearman rho correlations were used to assess collinearity between FRI-risk, FRI-fatigue and FAID-fatigue scores. Spearman’s rank-order correlation is a “Correlation coefficient indicating the magnitude of a relationship between variables measured on the ordinal scale.” (Polit & Beck, 2017). The correlations between the three models were strong. GLMMS examined the association between fatigue estimates and sickness absence accounting for the intraclass correlation coefficient (ICC) = 0.114. ORs and 95% CI were estimated. **All analyses were performed in STATA 14.1.
Nurses’ shiftwork characteristics, fatigue estimates and sickness absence.
Variables N (%)
Day 22 042 (53.06)
Night 19 496 (46.94)
0 to 40 (standard) 12 295 (29.97)
41 to 79 (medium-risk) 19 333 (47.13)
80 to 99 (high-risk) 6118 (14.92)
≥ 100 (very high risk) 3272 (7.98)
0 to 20 21323 (51.33)
21 to 40 1175 (2.83)
41 to 60 18561 (44.68)
≥ 61 479 (1.15)
M (SD) (range: 0.70–3.75) 1.12 (0.17)
Absent (yes) 2355 (5.37)
Present (no) 41 538 (94.63)
Note. *The sum does not add to 41 538 since the software considers the first seven days of the work schedule as history for cumulative fatigue.
Average FAID-fatigue score was 58.38 (SD = 26.63, range = 7-154), indicating fatigue to be a medium risk.
When FAID-fatigue risk scores were between 80-99, sickness absence was 1.63 times more likely than standard group
When FAID-fatigue risk scores were >=100, nurses were 1.73 times more likely to be absent compared to the standard group
The article showed tables with the FRI-fatigue and FRI-risk estimates but was complex and lengthy to place on this post. Summarizing the results:
FRI-fatigue scores >60, were 1.58 more likely to be absent compared to reference group
When FRI-risk scores increased by 1 point, the nurses were 1.87 more likely to be absent from work.
“This study provided strong evidence that high fatigue estimates are associated with sickness absence in 12-hour shift pediatric hospital nurses” and “strong correlations between nurses’ fatigue and risk estimates” were made. (Sagherian et al, 2017).
There was a lack of socio-demographic data and possible health issues that may have had an influence with absences and fatigue. Nurses may have worked additional jobs and family obligations were not taken into account in this study.
Out of 43,893 shifts, nurses reported for work 41,538 times and since the average FAID-fatigue score was 58.38 (which was medium risk), the vast majority of the nurses not only showed up for work, they were at medium risk for fatigue while they were there. Since it has been proven with higher fatigue comes higher risk for medical errors, an additional research study could build on this one looking into any medical errors that occurred during these 41,538 shifts. This study was already quite lengthy and a little complicated to comprehend at first, but if the researchers decided to continue with this thought process, it would give more concrete evidence supporting their original hypothesis.
Polit, D. F. and Beck, C. T. (2017). Nursing Research: Generating and assessing evidence for nursing practice (10th ed.). Philadelphia, PA: Wolters Kluwer.
Sagherian, K., Zhu, S., Storr, C., Hinds, P., Derickson, D., Geiger-Brown, J. (2017). Bio-mathematical fatigue models predict sickness absence in hospital nurses: An 18 months retrospective cohort study. Applied Ergonomics 73, pp 42-47. Elsevier.