Running head: GROUP CRITIQUE Group Critique of: Impact of a Nurse-Driven Mobility Protocol on Functional Decline in Hospitalized Older Adults Kristine Cargill, Cheryl Nienaber, Bailey Sundberg Ferris State University 1 GROUP CRITIQUE 2 Abstract The purpose of this paper is to take a quasi-experimental research study, Impact of a Nurse-Driven Mobility Protocol on Functional Decline in Hospitalized Older Adults (Padula et al., 2009), and critique it for the proper use of the 14 steps of the quantitative research process. The authors use Burns and Groves’ (2011) book, Understanding Nursing Research: Building an Evidence-based Practice, to support their reasoning of finding the study weak in all areas except for the authors’ purpose statement. Keywords: quantitative research, quasi-experimental study, problem, purpose, literature review, theory and conceptual framework, hypotheses, research design, sample, measurement and data collection, instruments, data analysis, study findings, study discussion, study conclusion GROUP CRITIQUE 3 Group Critique: Impact of a Nurse-Driven Mobility Protocol on Functional Decline in Hospitalized Older Adults Nursing research is the foundation for nursing practices. As defined by Burns and Grove (2011) “nursing research is a scientific process that validates and refines existing knowledge and generates new knowledge that directly and indirectly influences nursing practice” (p. 4). Solid, detailed research helps to develop, guide, and improve evidence based nursing practices. The research process of the article, Impact of a Nurse-Driven Mobility Protocol on Functional Decline in Hospitalized Older Adults (Padula, Hughes, & Baumhover, 2009) will be critiqued in this paper. The group used the textbook, Understanding Nursing Research: Building an Evidence-based Practice (Burns & Grove, 2011) to guide and support the critiquing process, while evaluating the strengths and weaknesses of the study. Purpose Statement Evidence This article focused on mobility issues of elderly patients related to hospitalization, with the purpose of the study “to determine the impact of a nurse-driven mobility protocol on functional decline in hospitalized older adults” (Padula, Hughes, & Baumhover, 2009, p. 326). For this study, participants were 60 or older, with a hospital stay of three or more days. Previous studies have documented the increase in dependencies of activities of daily living [ADL] and functional decline in older adults after hospitalization. The researchers hope to decrease the length of stay [LOS] of older patients and keep or increase their functionality with a mobility protocol (Padula et al., 2009, p. 325). GROUP CRITIQUE 4 Support Purpose statements describe the focus of the study, or why the study is being conducted (Burns & Grove, 2011). According to Burns and Grove (2011) “the goal of a study might be to identify, describe, or explain a situation; examine the effectiveness of an intervention, or determine outcomes of health care” (p. 146). Included in purpose statements are several additional components including: where the study takes place, the subjects being studied, and other variables. The purpose gives direction for the research study and without a purpose the audience would not know why a study is being conducted. Analysis The article clearly defines and includes the components of a purpose, making the purpose statement a strong section of the study. A clearly defined purpose statement informs the reader where the author is going with the study. The purpose statement in the article describes the components as follows: the population as older hospitalized adults, the independent variable as the mobility protocol and the dependent variable as the length of stay and functional status (Padula et al., 2009, p. 325). The authors’ purpose describes the goal of the study and how it may impact those involved. Problem Statement Evidence Research has shown a positive correlation between lack of mobility in acute care settings and increased LOS. The importance of achieving optimal health in order to promote continued independence, and the ability to maintain ADL are addressed (Padula et al., 2009, p. 325-326). While studies have shown that older adults who are a part of GROUP CRITIQUE early mobilization, physical, and occupational therapies have a higher rate of positive outcomes, evidence based guidelines have not been published (Padula et al., 2009, p.325). Support According to Burns and Grove (2011) “a research problem is an area of concern in which there is a gap in the knowledge base needed for nursing practice” (p.146). There are three distinct areas that need to be addressed in order to define the problem. First, what is the concern and why is it important to this population? Next, why is this significant to the population that is studied? And third, what is the current situation that needs to be studied and changed if relevant (Burns & Grove, 2007, p.146). Padula et al. (2009) state, "research has demonstrated that low mobility and bed rest are common during acute hospitalization" (p.1). Because of this research the authors hope that their study would, “lay the foundation for the development of standards of care" (p. 1). The authors support the significance of the problem by reminding the reader, "that given the majority of hospitalized patients in the United States are older than 65 years, this is a problem of great clinical significance" (p. 1). Burns and Grove (2011) state, “not all published studies include a clearly expressed problem, but the problem can usually be identified in the first page of the paper” (p. 146). According to Padula et al. (2009), “a stay in the hospital often results in complications that lead to functional decline” (p.1). While this information may be true, reasons for the increased complications and LOS are unknown. Padula et al., (2009) also goes on to state, “new functional deficits during hospitalization are also less likely to 5 GROUP CRITIQUE 6 recover lost function” (p. 326). The problem statement is vague, and the reader must uncover the true problem while reading through a majority of the first two pages. Analysis The problem statement is not strong in the groups’ opinion. Although the problem statement in the article and textbooks are similar, the article lacks concrete statistical information in the opening paragraphs to invite the reader to continue with the research. The article clearly states that the study is in regards to older population however it fails to list a range of participants at this point in the research. By reviewing the problem area of this article, it is possible that this would have been placed in the recycle bin. A clear problem statement would enhance the knowledge of the audience. While information is given throughout the article, the lack of interest would not have allowed for those seeking information to reach the target information. Literature Review Evidence Padula et al. (2009) used a variety of sources in what is assumed to be the review of literature section. The section is not clearly defined with its own heading, instead it is presented in the introduction section prior to the purpose statement. The authors listed 28 references at the end of their study, 16 of which are present in the assumed review of literature section. Five relevant primary sources were used. Upon research of the group these sources were found to consist of: one prospective multi-center study, one correlation research, two prospective cohort studies, and one quasi-experimental study. However, only two of the five primary and relevant sources were discussed in the narrative by name, and both lacking description as to the type of study done, and the year GROUP CRITIQUE done. These studies were completed by, “Brown CJ, Friedkin RJ, and Inouye SK (2004)” and “Brown CJ, Roth DL, Peel C, and Allman RM (2006)” (Padula, et. al, 2009, p. 331). The authors also included seven primary sources that were outside the five-year time frame to be considered relevant. The age range of these sources is from 1993-2003. Upon further research by the group into these seven sources it was found that all of them seem to have direct correlation with the purpose of this study based on their abstracts. Five of these sources were mentioned in the narrative by name, however, again lacking the year the study was completed. In some instances such as the description of study performed by Covinsky KE, Palmer RM, Fortinsky RH, et al (2003), the authors included that this was a “descriptive study of 2293 older adults’ ADL function during an acute hospital stay, found that 35% of patents had worse ADL function at discharge compared with preadmission baseline” (Padula, et al., 2009, p. 326). The final four sources used by Padula, et al. (2009) were secondary sources, two of which upon further investigation by group, and based on abstracts listed in the CINAHL database had no direct relationship to the purpose of their study. Only one of the secondary sources were cited in the narrative by their names, Callen BL, Mahoney JE, Wells TJ, Enloe M, and Hughes S. (2004) with a brief mention that Callen, et al, “found little research specific to mobility changes during hospitalization” (Padula, et al., 2009, p. 326). Support Burns and Grove (2011) state, “a review of literature provides you with the current theoretical and scientific knowledge about a particular problem, enabling you to synthesize what is known and not known” (p. 189). After the authors of a study have 7 GROUP CRITIQUE stated their problem statement the literature review should then provide the reader with three concepts: “a description of the current knowledge”, “identification of the gaps in this knowledge base”, and “the contribution of the present study to the building of knowledge in this area” (p.189). However, it is important to note that whether primary or secondary sources are used, the source must be relevant. Burns and Grove (2011) define relevant sources as those “with direct bearing on the problem of concern…and published within five years of the acceptance of the manuscript for publication” (p.190). Finally the difference between primary sources and secondary sources is the author. In a primary source the author is also the one who completed the study or research, and thus provides any researcher with highly credible information regarding the topic (Burns & Grove, 2011, p. 191-92). A secondary source “summarizes or quotes content from primary sources” which could result in the authors’ works being interpreted differently (Burns & Grove, 2011, p.192). Analysis The Review of Literature [ROL] section was not labeled, and actually based on assumption of the section considered review of literature preceded the purpose statement. The ROL section was comprehensive in the fact that it covered much material, however, not concise or even cohesive. It was difficult at times to determine why this information presented was important to the article. This could have been relieved by having the ROL section follow the purpose statement and clearly identified with a heading. Further, as previously assessed by group the problem statement in this article was unclear, and based on ROL section it remains unclear. The authors used many sources: five primary sources that were less than five years old at time of publication, seven primary sources greater 8 GROUP CRITIQUE than five years old, including two studies that could be considered landmark sources. Finally there were four secondary sources, two of which had no bearing on topic of this article. Many of the secondary sources were used only for their statistical information, however, all information was paraphrased and the authors also paraphrased several secondary sources at one time without clearly stating what information came from what source. Throughout the entire assumed ROL section the authors jumped from sources, and then would return to sources in another paragraph. This added to the confusion for the reader trying to keep sources strait. It would have been better if the authors stuck to one source at a time and critiqued/summarized all of it before moving to the next. Further, the sources were rarely introduced with the author’s names or the year of publication leaving the reader to flip from content to reference page through the entire assumed ROL section. Of the eight studies called out by name by the authors, seven were primary sources, however only two sources were within the five-year relevant time frame. However the other five were credible, significant studies that had direct bearing on the problem and purpose of this article. Having primary sources greater than five years old shows that this problem has been researched before and has significant influence on obtained evidence based practice guidelines for the formation of mobility protocols in the elderly hospitalized for acute conditions. In conclusion it is the opinion of the group that had the study of Padula et al. (2009) should have clearly stated their ROL section with a proper heading that followed the purpose statement. The authors also should have selected four or five primary 9 GROUP CRITIQUE 10 sources, both within the five-year time frame and those that are considered landmark sources to fully describe and critique in this section. The authors should have been listed in the narrative of the ROL section, with the year of publication clearly stated as well. By reviewing and appraising one source at a time and thoroughly would have helped flow this section logically and enhanced the importance of the problem being discussed. Theory and Conceptual Framework Evidence The article clearly stated that model and training were provided for staff members. Evaluation and Specific Interventions for Successful Healing [GENESIS] “includes a nurse-driven mobility protocol that served as the intervention for this research” (Padula et al., 2009, p. 328). This study focuses on elderly patients and the framework appears to be appropriate as the GENESIS model includes a mobility protocol which states “nurse is directed to question orders for bed rest as well as to routinely evaluate the necessity of obstacles to mobility such as urinary catheters (Padula et al., 2009, p. 328). The GENESIS model is suitable to measure the stated goals of the study. Padula et al., (2009) also used other measurements in the study. The Get Up and Go test provides an objective assessment with criteria that can be measured not only for this study but also for future studies. The mini mental exam is used to evaluate patients, however, “the significant other’s perception of the patient’s abilities was assessed” (Padula et al., 2009, p. 329). The final measurement tool was the Barthel Index (BI). This test measures the subjective data in relationship to ten basic ADL’s. Staff training included four hours of education including the importance of adhering to the mobility protocol. Ambulation, mealtime activities, restroom concerns, GROUP CRITIQUE and barriers were reviewed with staff members prior to the start of the study. The integrity of the research is strengthened by the clarification that “treatment and control groups were on different floors of the hospital, and nurses did not float between the units, minimizing the chance of intervention affecting the control unit” (Padula et al., 2009, p. 328). Although the sample population selected was a convenience sample, stringent variables are used to strengthen the overall study. The Mini-Mental Status Exam was one variable used to measure the ability of the patient to participate in the study. Other variables included the length of stay, and the ability to understand (Padula et al., p. 327). The concepts are clearly defined in the research. The audience is able to understand the connection between the variable and why each one was chosen. The research must be presented in such a way that removes potential barriers, such as non-English speaking patients. It is assumed that the population of the site, Miriam Hospital, is a facility that predominately speaks English. The variables stated are “the independent variable was mobility protocol; dependent variables were functional status and length of stay” (Padula et al., 2009, p. 327). Propositional statements that are used to guide the research include the hypothesis statement, “(1) older adults who participate in a mobility protocol will maintain or improve functional status from admission to discharge and (2) older adults who participate in a mobility protocol will have a reduced LOS” (Padula et al., 2009, p. 327). Support When reviewing an article to determine the worthiness of the content, a variety of thoughts must be considered. A theory is “the initial inspiration for developing a study” 11 GROUP CRITIQUE (Burns & Grove, 2011, p.227). However, what lead to the inspiration of the theory is the quest for additional knowledge. As concepts are considered, the researcher pulls together information that seems to support the phenomena to attempt to “describe, explain, predict, and control” the expected outcome (Burns & Grove, 2011, p. 228). Burns & Grove (2011) states that a “framework is a brief explanation of a theory or those portions of a theory that are to be tested in a study” (p. 227). A researcher uses the framework to build a strong foundation for the study; a stepping-stone to expand on the current knowledge base. The framework of the theory has four key elements. Each of the four areas are interwoven to produce the strength of the theory. Burns & Grove (2011) listed “concepts, definitions of concepts, relational statements, and map or model” as the elements used to understand theories (p. 230). Concepts provide the audience with the big picture, an overall basis for the theory, thus a stronger concept will provide a stronger theory. While it is possible to use the terms construct to give an overall view, the term variable is used to strengthen the validity of the research due to its narrow definition (Burns & Grove, 2011, p. 230). Analysis While some may view this section as a strong, others will not. It is clear that the background of the audience is needed to validate the strength of this research article. While it is clearly stated that the GENESIS model is used as a framework for this study, some audience members may not have heard of this model. BI scores are compared to Frail Elderly Functional Assessment, and while this is meaningful in the study, the scores are subjective, and thus could be dependent on other factors not considered in the study. The Mini-Mental State Examination measures the significant other’s perception; again, 12 GROUP CRITIQUE 13 this is not a clear-cut assessment tool. Outside factors such as optimism, faith and expectations play into the measurements. Scores for the Up and Go test are defined and measurable and can be reproduced in future studies, making this a strong assessment tool. Audience members who are unfamiliar to any of these measurement tools will have difficulty understanding the importance of their use in this study. While all of these tools are clearly operational, since they are associated with numbers, the lack of understanding how these numbers are achieved provides additional frustration to the audience. Thus leading to confusion and lack of clarity as to why the researcher chose these particular instruments and this model as a part of the framework. The goal of this article is to provide information that can be used to validate current practice, or to review existing protocols and make adjustments to ensure the best outcomes for our patients. As an audience member, the pure effort needed to examine the term GENESIS, would leave questions regarding other possible terms used that would need to be researched. The purpose of a study is to help the audience comprehend the information provided, not to cause additional work. While the intent of the researcher is positive, the lack of background information of the research tools leaves a negative tone to the audience. Therefore, this area of the article must be viewed as a weak area. If the audience is lost, the results are meaningless. Hypotheses Evidence The article states two hypotheses, “older adults who participate in a mobility protocol will maintain or improve functional status from admission to discharge and, older adults who participate in a mobility protocol will have a reduced LOS” (Padula, et GROUP CRITIQUE al., 2009, p. 327). The variables listed being, “the independent variable was mobility protocol and dependent variables were functional status and LOS” (p.327). Support: According to Burns and Grove (2011) the “hypothesis is a formal statement of the expected relationship(s) between two or more variables in a specified population” (p. 167). A study can have any number of hypotheses that “translate the research problem and purpose into a clear explanation or prediction of the expected results” (p.167). Hypotheses are used to predict the outcome of the study; therefore, the purpose of the study dictates the type of hypotheses used. There are four categories of hypotheses, “associative verses casual, simple versus complex, non-directional versus directional and, null versus research” (p. 167). An associative hypothesis “proposes a relationship among variables that occur or exist together in the real world”, but do not state that “one variable causes an effect on another variable” (p. 167). However, casual hypotheses do “propose a cause-and-effect interaction between two or more variables” (p.170). The difference between simple and complex hypotheses are the number of variables present in the study, two variables means the hypothesis is simple, three or more means it is complex (p. 172). A non-directional hypothesis is defined as one that “states that a relationship exists but does not predict the nature of the relationship” (p. 173), where as a directional hypothesis does predict the nature of the “interaction between two or more variables” as either positive or negative (p. 174). Finally, a null hypothesis is “used for statistical testing and for interpreting statistical outcomes”, it essentially states that the researchers expect “no relationship between two variables” (p. 174). The opposite of the null 14 GROUP CRITIQUE hypothesis is the research hypothesis, which will “state that a relationship does exist between two or more variables” (p. 175). Analysis: This study clearly identifies two hypotheses immediately following the purpose statement. Both statements are clear, concise, and declarative as evidenced by including the word “will” (Padula et al., 2009, p. 327). The group believed that the problem statement section of this study was not strong, however, the key identified or assumed variables to the problem were lack of mobility protocols, decreasing functional status in patients, and longer lengths of stay resulting from decreased functional status. Therefore, the group agrees that these stated hypotheses are directly tied to the study problem. As also identified by the group, the framework for this study the GENESIS program for geriatrics, contains an age-specific mobility protocol and set of interventions, therefore with the framework being a mobility protocol it can be said that both hypotheses are derived from the framework. Both of the hypotheses included the population to be “older adults”, however, should have included the exact parameters of older adults that was not directly indicated until discussion about the sample. Further, neither of the hypotheses indicated directly the independent or dependent variables until research design section following the stated hypothesis. It can be assumed the variables by the reader in the hypotheses statements, however, if you are going to directly label them why not do so with the hypotheses themselves. Both of the stated hypotheses are casual, complex, and directional. They both state a “cause-and-effect relationship between two or more variables” (Burns & Grove, 2011, p. 170), contains a total of three variables, and also they both state the exact nature 15 GROUP CRITIQUE 16 of the relationship (p. 174). The hypotheses are research hypotheses, therefore, the null hypothesis is implied to be the exact opposite of the hypotheses, thus that mobility protocols do not have any effect on functional status or length of stay in older adults. Finally, both hypotheses are testable which is discussed in the next section of research design. Research Design Evidence The authors of this article describe the research design as a nonequivalent control group design. The authors also list the independent variable as the mobility protocol and two dependent variables functional status and length of stay (Padula et al., 2009, p. 327). Support According to Padula et al. (2009), “a research design is a blueprint for conducting a study” (p. 253), which help control items that would impede validity. The more control a researcher has over the study the more valid the study becomes and research designs help to provide control over the study. Further, controlling variables and studies helps reduce biases (p. 253). There are many different types of experimental designs. A few examples of research designs are: quasi-experimental design, experimental, and non-experimental research design. Quasi-experimental design is a study design about cause and effect. This design “facilitates the search for knowledge and examination of causality in situations in which complete control is not possible” (Padula et al., 2009, p. 270). Comparison groups are used in this design and participants are randomly assigned to the experimental or comparison group. Experimental research designs can vary but all focus on casualty or GROUP CRITIQUE cause and effect. Non-experimental research designs do not control variables, subjects, or conditions. Nonequivalent control group design is a form of quasi-experimental research design. This design uses pretest and protest where one group receives a specific treatment and the comparison group does not receive the specific treatment, however, the comparison group does receive a standard of care. Subjects are chosen from the same group of possible subjects and then choose whether to receive treatment or not to receive treatment. This research design does not randomly select the comparison group. “These groups cannot be considered equivalent, however, because the subjects in the comparison group usually differ in important ways from those in the treatment group” (Padula et al., 2009, p. 271). Analysis The authors did have a designated section with a heading for the research design method; however, this section was weak. The author used a nonequivalent control group design which means the experimental group and the comparison groups were not equal. Internal validity is threatened in this study because the groups were not equal. The authors believed that participants that followed the mobility protocol would have improved functional status and length of stay in the hospital. If the groups were not equal, other factors may have contributed to the outcome of the study. The author did acknowledge that other factors may have influenced the outcome. If the groups were not equal, then the audience may have doubts that the mobility protocol affected the length of stay and the functional status results. Also, the audience may question whether the same results would occur in a similar study. 17 GROUP CRITIQUE 18 Sample Evidence The article clearly identifies the target population under the sample section, as adults “age 60 and older, and admitted with a medical diagnosis to one of two nursing units (n=25 each)” (Padula et al., 2009, p. 327). While the hospital population is obviously much larger, the researcher chose to use a convenience sample from within these two units as subjects for the study. As with any reliable study the researcher described the sampling method used to select participants. Functional status, “LOS, ability to understand English, no physical impairment, and cognitive function all were part of the inclusion criteria” (Padula et al., 2009, p.327). The exclusion of subjects was based on a score below 24 on the Mini Mental Status Examination, transfer from the unit prior to the end of the study, and a LOS under three days (Padula et al., 2009, p. 327). A total of 84 subjects were enrolled into the study for those that were eligible; 34 withdrew for a variety of reasons. The attribution rate for the study was 17% and the retention rate was 100%. “Sample size statistics were conducted to determine adequate power and supported a total number of 50” (Padula et al., 2009, p. 327). The researcher utilized two homogeneous units related to “size, patient population, and unit staff composition” (Padula et al., 2009, p. 327). Support A researcher would find it nearly impossible to study all individuals in order to create a theory. Instead, the researcher will select a sample; “sampling involves selecting a group of people, events, behaviors, or other elements with which to conduct a study” (Burns & Grove, 2011, p.290). By using specific criteria the researcher will be able to GROUP CRITIQUE focus on the individuals that will provide evidence that the theory is correct, or prove that not enough sufficient evidence exists to confirm the hypothesis of the theory. “If a study is of high quality with finding that are consistent with previous research, then the researchers can be confident in generalizing their finding to the target population” (Burns & Grove, 2011, p.291). Many steps are necessary to ensure that the sample is representative of the population. To start, we can use the example in Burns and Grove (2011) regarding diabetes; the target population would consist of all diabetics (p. 290). With the target population identified it now is necessary to determine who “the portion of the target population to which the researcher has reasonable access” (Burns & Grove, 2011, p.290). Next, is to determine the inclusion, what makes the person part of the population that the researcher wants to study, or the exclusion criteria, what characteristic does the individual possess that may pose a threat to the validity of the study. The objective of the researcher is to acquire homogeneous subjects. The similarity of the subjects will provide a basis for the study thus allowing the researcher to make a generalization when complete (Burns & Grove, 2011, p. 290). The sample must be representative of the population. Researchers conduct studies using a variety of methods dependent on the goal of the study. “A simple random sample is the most basic of the probability plans. It is achieved by randomly selecting elements for the sample frame” (Burns & Grove, 2011, p.299). This process can be thought of as a raffle, each person eligible is considered to have an equal chance at being selected by pure chance. Stratified random samples are groups from within a larger group. For example a high school may contain four levels of students: freshmen, sophomores, 19 GROUP CRITIQUE juniors, and seniors. To make sure each group is appropriately represented, the same number of subjects will be chosen from each grade level. Cluster sampling uses a macromicro approach; researchers start with the largest group possible and then narrowing the focus to make a generalization regarding the group (Burns & Grove, 2011, p. 299). Burns and Grove (2011) also list systematic sampling, quota sampling, and nonprobability sampling as ways to obtain subjects for a study. However, one of the weakest types of sampling is the convenience sample. Subjects are chosen purely because they are in the “right place at the right time” (p. 305). Researchers will often use convenience samples due to low cost and easy accessibly of the subjects. They feel that in the healthcare field it is often better to use a convenience sample thus increasing the sample size by including all of the potential subjects who meet the specified criteria (Burns & Grove, 2011, p. 305). Sample bias is a major concern with convenience samples. “Bias in a study distorts the findings from what the results would have been” (Burns & Grove, 2011, p. 254). However, by recognizing this as a potential complication, researchers can design the study to limit the number biases and therefore create a stronger foundation for the study. While not all biases are evident, by critically evaluating study methods, treatment, and tools, the biases can be recognized and reported in the study (Burns & Grove, 2011, p. 254). Analysis The researcher listed the target population for the study and attempted to use inclusion and exclusion criteria that would make both units as homogeneous as possible. While this effort did strengthen the overall point of the study, there are multiple variables 20 GROUP CRITIQUE that were not addressed. Criteria that should be reviewed include the random selection of the patients in the study. How would the results differ if the subjects were randomly divided into equal groups on each floor instead of having one floor as a treatment group and the other as no treatment? The sample size (n = 25 each unit) is based from a convenience sample; the relatively small size can easily cause bias within the results. Although the adequate power of the study does support the size, in reality, the study design is still weak due to the convenience sample method. The mobility of each patient is not only based on the efforts of the health care staff but also the efforts of the individual patient. The mindset of the individual will aid the staff to facilitate increased mobility providing an enhanced outcome to the client. Social factors are not considered within the study framework. Family, friends, neighbors, groups, and organizations are all key components to be considered. The focus of the study is on seniors’ age 60 and older, the hospital setting may be the only social contact for the patient; this lack of social support may decrease the initiative of the patient to participate with the mobility protocol. Again, we do not know if this was a concern or if there were equal support systems in place on both units. Overall health and co-morbidities were not discussed. As an audience we are unaware if this was addressed, and if so, how similar were the groups? The characteristics of the nursing staff are provided for each unit yet it fails to mention the experience level of each unit. While the nurses did not float between floors to work, the opportunity still existed for staff to share experiences and ideas (Padula et. al, 2009, p. 327). These questions leave the audience to ponder if the treatment was successful or if other factors came into play and made the difference between the treatment and control groups. 21 GROUP CRITIQUE 22 Measurement and Data Collection Evidence For this study, an advance practice nurse was hired to collect data. The nurse had experience in the geriatric field and was also trained by the investigator and a geriatric clinical nurse. Padula et al. (2009) state that, “training included human subjects’ protection and achievement of high-level proficiency with the protocol and data collection instruments” (p. 328). The research nurse screened all patients using the study instrument, which was tested for reliability prior to this study until an acceptable level was reached. If patients met eligibility requirements, they received an information letter regarding the study (p. 328). A data collection sheet was used to collect data for this study. The tool was used to gather data such as: “age, gender, primary diagnosis, use of assistive devices, fall risk assessment, presence of any restrictions to mobility (eg. bed rest order, foley, oxygen use, restraints), use of occupational or physical therapy, LOS, first and number of times out of bed, and type of bed activity” (Padula et al., 2009, p. 328). The other tools used in this study were the Get Up and Go test and the BI. These tools will be described in further detail in the instrument section below. The data was collected at Miriam Hospital, an acute care hospital, on two separate but similar units in size, patient population, and nursing staff. The collection process occurred within 48 hours of admission; however, for the BI instrument, self-perception of functional status was measured at two weeks before admission, at admission, then at discharge (Padula et al., 2009, p. 328). GROUP CRITIQUE Support According to Burns and Grove (2011), “data collection is the process of acquiring the subjects and collecting the data for the study” (p. 361). The data collection process should be detailed in the research study and the steps should be clear, although different with each study depending on the design and measurement techniques. The steps or process of data collection consists of five tasks completed simultaneously: “selecting subjects, collecting data in a consistent way, maintaining controls, protecting the studies integrity, and solving problems that threaten the study” (Burns & Grove, 2011, p. 362). The process of selecting subjects that meet criteria, descriptions of those that decline participation, measurements, and time and setting of measurement should be included in the report as well (Burns & Grove, 2011, p. 362). “Measurement is the process of assigning numbers or values to individuals’ health status, objects, events, or situations using a set of rules” (Burns & Grove, 2011, p. 327). There are four different levels of measurement: nominal, ordinals, interval, and ratio. Each level of measurement has a statistical test that is used. According to Burns and Grove (2011), nominal-scale measurement is used when subjects or objects will be placed “into categories of a defined property”, such as gallbladder stone and cellulitis (p. 329). Three rules exist for categories in nominal measurement; they must be exclusive, exhaustive, and not orderable (Burns & Grove, 2011, p. 329). Ordinal-scale measurement must also be exclusive and exhaustive; however, the categories can be ranked (Burns & Grove, 2011, p. 328). The category intervals are unequal, meaning there may be a difference between the rankings. The interval-scale measurement has numerically equal intervals (Burns & Grove, 2011, p. 329). The 23 GROUP CRITIQUE categories in this scale are exclusive, exhaustive, and orderable. This scale lacks a zero point and represents a range of values. The ratio-scale measurement “meets all of the rules of other forms of measurement: mutually exclusive categories, exhaustive categories, ordered ranks, equally spaced intervals, and a continuum of values” (Burns & Grove, 2011, p. 330). This is considered the highest level of measurement and has a zero point. Ratios help determine the differences among groups and determine the strength of relationship between the groups. Two items that may affect data collection are Hawthorne effect and researcher effect. The Hawthorne effect occurs when subjects act in a certain way because they know they are in a study. This effect may cause unintended outcomes to studies or experiments. The researcher effect occurs when the researcher is bias or influences certain behaviors of subjects in the study creating conditions that are not equal or identical. Avoiding these effects allows for increased validity of the study. Analysis The data collection section was not detailed thoroughly by the authors, making this section weak. The information regarding data collection was spread throughout the article and not confined to its own section. The data collection is suppose to answer the: who, what, where, when, and why of a study. In this study the authors did not include the length of time it took to collect the data for all participants. The article did state that data was collected within 48 hours of admission and a time frame for the BI instrument was provided (Padula et al., 2009). The authors stated who collected the data, the advance practice nurse, and described some of her background to let the audience know the 24 GROUP CRITIQUE 25 collector had experience with geriatrics. The data was collected on two units but the authors did not describe what type of units they were and the type of unit could potentially affect the outcome; however, the authors did state the two units were similar. The authors did not discuss why the data they collected was collected; however, a description of what data was collected was listed as described above such as fall risk assessment. The level of measurement was not provided the audience, leaving questions regarding the statistical analysis of this study. By not providing the audience with the measures, the audience does not know if the study is using the correct statistical tests. This is a threat to internal validity. Also, the audience does not know if Hawthorne or researchers effects affected the data collection in this study, which would be a threat to external validity. In return, the audience is left to question whether the outcome is affected by Mobility Protocol or the how the data was collected. Instruments Evidence The authors began by using a data collection sheet, developed specifically for this research and included such information as age; gender; primary diagnosis; use of assistive devices; fall risk assessment; presence of any restrictions to use of occupational or physical therapy; LOS; first and number of times out of bed; and type of outof-bed activity (Padula et al., 2009, p. 328). GROUP CRITIQUE The authors determined functional status using “the modified BI and the Up and Go Test” (Padula et al., 2009, p. 328). Compared to the original BI, “the modified version used, retained the original 10 items but included a 5-point rating scale for each item to improve sensitivity to detecting change” (Padula et al., 2009, p.328). Individual test results will range from, “0 (totally dependent) to 100 (totally independent)” (Padula et al., 2009, p.328). The BI is taking the place of the traditional assessment of Frail Elderly Functional assessment as “researchers have proposed the BI as the standard for clinical and research purposes” (Padula et al., 2009, p. 328). The two were shown to exhibit “correlations of … 0.91…the interrater agreement of r = 0.793 was reported for the overall BI score” (Padula et al., 2009, p. 328). The second “objective assessment” tool, was the Up and Go Test. The authors stated. “the get up and go test has been shown to be reliable and valid and is reported to correlate with the BI at r = -0.78” (Padula et al., 2009, p. 328). Test scores for the Get Up and Go Test are given “on a 1 to 4 scale, 1 being “able to rise in a single movement and 4 being unable to rise without assistance” (Padula et al., 2009, p. 328). Finally the authors measured cognitive status using the “mini-mental state examination” without offering any information on reliability and validity. Participants who “score[d] less than 24, the significant other’s perception of the patient’s abilities [were assessed]” (Padula et al., 2009, p. 328). Support According to Burns and Grove (2012), reliability deals with the “consistency of the measurement method” and reliability testing is the “measure of the amount of random error in the measurement technique” (p. 332-333). There are three areas that reliability 26 GROUP CRITIQUE testing focus on, “stability, equivalence, and homogeneity” (Burns & Grove, 2012, p. 333). An instruments’ stability or test-retest ability “is concerned with the consistency of repeated measures, of the same attribute, with the use of the same scale or instrument” (Burns & Grove, 2012, p. 333). There is also interrater reliability, which measures the equivalence between “two versions of the same-pencil-and paper instrument or of two separate observers” (Burns & Grove, 2012, p. 333). There is not a specific value at which “interrater reliability is unacceptable, however, any value below 0.80 should generate serious concern about the reliability of the data or of the data gathered” (Burns & Grove, 2012, p. 333). What the interrater value means for a value of 0.80 is that there is 80% reliability and 20% random error, obviously the less the random error the more reliable the instrument. Finally, homogeneity is the final type of reliability testing used “primarily with paper-and-pencil instruments or scales to address the correlation of each question” with the others in the instrument (Burns & Grove, 2012, p. 334). Also known as a measure of the internal consistency, reliability of 0.80 is considered strong for an instrument that has been used multiple times in the past, however, a new instrument having a reliability of 0.70 is considered acceptable. Again, the higher the reliability, the less chance there is of random error. Validity of the instruments is a measurement of how well the instrument “reflects the abstract concept being examined” (Burns & Grove, 2012, p. 334). According the Burns and Grove (2012), “no instrument is completely valid…validity will vary from one sample to another and one situation to another” (p. 334). Evidence of validity are presented in published studies in three common ways: “contrasting groups, convergence, 27 GROUP CRITIQUE and divergence” (Burns & Grove, 2012, p. 335). Evidence of validity from contrasting groups is the tested when the same instrument is completed by two separate groups “known to have contrasting scores” (Burns & Grove, 2012, p.335). Evidence of validity from convergence deals with a new instrument being compared with an existing instrument, where as evidence of validity from divergence deals with two instruments including one that measure the direct opposite of the concept under study (Burns & Grove, 2012, p. 335). For example: “if the newly developed instrument is a measure of hope, you could make a search for an instrument that measures despair…if the divergent measure (despair) is negatively correlated with the other instrument (hope) that validity of both instruments is strengthened” (Burns & Grove, 2012, p. 335). Analysis The instrument section of this published study is clearly labeled, and consists of three separate instruments: BI, Up and Go Test, and Mini-mental state examination. The authors clearly described the BI test, how and when it was administered, and that this instrument measured the self-perceived ADL functional ability of the patient. The authors mention the Frail Elderly Functional Assessment and provide evidence of validity from convergence of 0.91, which based on the text increases the validity of the new instrument of BI. However, the evidence of validity cited in the study is cited from another study, indicating that the authors themselves did not verify the validity of the BI. Further the BI and the Frail Elderly Functional Assessment are not equal instruments: as the BI is purely subjective and the Frail Elderly Functional Assessment is both subjective and objective, nor did the authors state the test of validity. Further while the BI was mentioned to be the new standard, the authors neglected to include this in the narration part, and only cited 28 GROUP CRITIQUE their source. The type of reliability testing done for the BI was not concretely stated, it leaves the readers to assume the authors chose equivalence or interrater reliability which resulted in a reading of r= 0.793, which is .007 below the unacceptable value. The instrument, Get Up and Go, the authors stated that it was reliable and valid, however, based on another study’s conclusions, appearing to the reader that the authors did not test the reliability and validity themselves. According the study cited, the BI and Get Up and Go test correlate at r = -0.78 (p. 328). The authors did not express the type of validity completed on the two tests, leaving the reader to individually determine that it was from divergence, based on the fact the authors described this test as objective. According to the text having negative correlation strengthens both instruments. Nothing related to reliability or validity was mentioned about the third instrument used, the minimental state examination, borrowed from another study or completed by the authors themselves. There is a huge threat to internal validity in this study based on instrumentation. First, the levels of reliability and validity of both the BI, and Get Up and Go test were cited from another study, no information was given to conclude that it was even measurement on the same type of sample. Also the BI instrument is a pencil and paper instrument given at admission inquiring about two weeks prior to admission, at admission for present level of mobility, and again at discharge. Therefore, the answers achieved are in question, and results invalid because the question arises as to whether taking the same test more than once produces real results or results based on seeing the same questions before. 29 GROUP CRITIQUE 30 Data Analysis (Descriptive Statistics) Evidence The article includes frequency data regarding the research. Ungrouped frequency revealed the use of 23 men and 27 women in the study, and also the number of participants that used assistive devices prior to admission. Occupational and physical therapy participation is noted along with use of restraints. Grouped frequencies list where the patients were living prior to admission. Measures of central tendency include the mean age of 80.4 and 6.7 as the mean number of diagnoses. Measurements of dispersion utilize the age range of 62 to 97 years old. Standardized scores were not utilized in this study as the groups were evenly divided between the treatment and the control group (Padula et al., 2009). Table one provides various information regarding the two units that were used for the study. Support Researchers need to help the target audience understand the population that was used, and which measurements were utilized to explain variables present in the study. Four key areas to understand are “frequency distribution, measures of central tendency, measures of dispersion, and standardized scores” (Burns & Grove, 2011, p. 383). Frequency distribution is divided into two classifications: grouped and ungrouped. While ungrouped provides data that is discrete such as sex, gender, ethnicity, ungrouped distribution is used for continuous information like age, weight, and time (Burns & Grove, 2011, p. 384). Measures of central tendency include the mode, median, and mean. While each of these represents a point of information, each does so in its own unique way. The mode simply is the “numerical value or score that occurs with the greatest GROUP CRITIQUE frequency” (Burns & Grove, 2011, p. 385). To find the median the data is placed in numerical order and the number in the middle is the median. If there are an odd number of items, then the two numbers in the center are added together and divided by two to obtain the median. The mean is the “sum of the scores divided by the number of scores being summed” (Burns & Grove, 2011, p. 387). Measures of dispersion allow the researchers to provide information that cannot be obtained from the mode, median, or mean. While the range of a measurement is simple to figure out, it is sensitive to numbers that are extreme in one direction or another. A “variance value has no absolute value and can be compared only with data obtained using similar measures” (Burns & Grove, 2011, p. 388). The standard deviation is used to help understand the relationship of the data. In a normal bell shaped curve “68% of the values will be within one standard deviation above or below the mean, 95% will be within 1.96 standard deviations above or below the mean, and 99% will be within 2.58” (Burns & Grove, 2011, p. 388). Standardization of scores is necessary in order to make accurate comparisons. Raw scores are calculated and given a Z-score. Thus, when comparing two groups that are unequal in number, the researcher can use a Z-score to measure both the uniqueness and the similarities of groups (Burns & Grove, 2011, p. 388). A scatterplot is frequently used to show a graphic view of the data. By placing information in the graph researchers can look for patters and often reveal “the degree of relationships between the variables” (Burns & Grove, 2011, p. 389). 31 GROUP CRITIQUE 32 Analysis The article does provide basic information to begin to understand the groups that were used in this study. However, a table listing each participant’s age, diagnosis, LOS, BI score, Up and Go score and other variables would provide a better picture for readers to grasp both the commonalities and the distinct differences between the groups. It is important that the researcher noted that the groups are comparable, yet to strengthen this claim a chart would provide additional information at a glance. Slightly over 50% of the participants had either occupation therapy or physical therapy ordered. Of this group (which the author again states is approximately equal) how did this effect the overall outcome for the patient? Was it those who received this therapy that scored the highest gains in mobility, or those that were assisted only by nursing staff? How is it possible to compare this group with those not receiving therapy and yet claim that mobility by the nursing staff increases patient outcomes? Unfortunately, the descriptive statistics leaves the audience with more questions than answers. The author might consider adding a chart to future studies to strengthen the information provided for easier understanding of the audience. Data Analysis (Inferential Statistics) Evidence Fall risk scores for subjects prior to admission was scored at P=.07. Table two uses Barthel scores for preadmission, admission, and discharge. Improvement in discharge scores from preadmissions, P=.05 and decrease in discharge scores from preadmission, P=.006. Other scores include out of bed to chair, P=.05; ambulation in the hallway, P=.007; LOS P<0.001 (Padula et al., 2009, p. 329). GROUP CRITIQUE Support The goal of the researcher when selecting a population is to gain as much information as possible to support the given claim. From there “an inference is made from the study findings obtained from a specific sample and applied to a more general public” (Burns & Grove, 2011, p. 378). A t-test is used to “test for significant differences between two samples” (Burns & Grove, 2011, p. 404). While it is possible to encounter a type I error if this test is used incorrectly, the t-test remains general accepted unit of measurement. A t-score can be used for relatively small sample sizes and provides a slightly wider confidence interval than a z-score. Analysis The author does not do a very good job of presenting the information here. There is very limited statistical information to make inferences. These few statistics provided are meant to assist us in deciding if there is enough evidence to pursue further research in this area, or if there is not enough evidence to support the claim. The limited size of the study (n=25) in each unit, along with the P-value leave room for errors in the study. An alpha score should be utilized in this study as this article pertains to nursing which tends to operate on an alpha = 0.05. While it is not useful to make the criteria extremely limited, the author does not place enough information in here to fully appreciate the potential for increasing mobility to provide better functioning. More testing and measurements need to be evaluated in order to consider this study truly helpful. 33 GROUP CRITIQUE 34 Study Findings Evidence The study was to research two separate hypotheses. First, “older adults who participate in a mobility protocol will maintain or improve functional status from admission to discharge” (Padula et al., 2009, p. 327). The authors indicate that this was supported through their results. The researchers present results indicating the BI score of the treatment group had greater improvements over the control group (P=0.05). Hypothesis two states “older adults who participate in a mobility protocol will have a reduced LOS” (Padula et al., 2009, p. 327). This hypothesis is also supported (P <.001) according to the authors. Support Findings from a study represent the information gained by researchers through testing. As evidence is evaluated, the researchers begin the process of exploring the significance of the findings. “Significance may be associated with the amount of variance explained, the degree of control in the study design to eliminate unexplained variance, or the ability to detect statistically significant differences” (Burns & Grove, 2011, p. 410). Some studies are clearly more important than others, and several may not be important for years. “The strongest findings of a study are those that have both statistical significance and clinical importance” (Burns & Grove, 2011, p. 410). Nursing research seeks to find meaning form the study. “Implications are based on but more specific than conclusions, and they provide specific suggestions for implementing the findings in nursing” (Burns & Grove, 2011, p. 414). Nurses, as well as the entire medical community, are constantly striving to improve current practices with GROUP CRITIQUE 35 the best evidence available. It is important to stay current and continue to develop research to best serve patients and the community. Analysis This section is quite confusing to the audience. Common sense would have validated the same results. If a patient is mobilized they will have better functional ability; a patient that is active will have a shorted LOS. This is basic information that the audience already knows. The researchers fail to explain why this information is important and what difference it will make in current treatment plans. As health care providers basic goals are not changed due to these statistics. Members of the treatment team will still assist patients in mobilization as it is good practice. This section is weak. Other than providing information that should already be practiced, nothing is new to the audience. Study Discussion Evidence According to Padula et al., (2011) “A significant decline in function between preadmission and admission was detected in both groups, a finding with important implications for both lay and professional care providers” (p. 330). Limitations of the study include that groups were unequal at the start and other factors may have come into play that are not reported. The researchers’ so suggest that additional studies “with quantification of the impact of diseases is indicated, especially because participants in the study experienced on average, about seven comorbid conditions” (Padula et al., 2011, p. 330). GROUP CRITIQUE Support This area of the research is where evidence is pulled together to give them significance. “This section includes major findings, limitations of the study, conclusions drawn from the findings, implications of the findings for nursing, and recommendations for further research” (Burns & Grove, 2011, p. 59). Limitations need to be reviewed so the audience can fully understand the importance of the information provided. “Limitations are restrictions in a study that may decrease the credibility and generalizability of the findings” (Burns & Grove, 2011, p. 48). By relating the findings to the framework provided in the study, researchers will validate the findings and increase the validity of the research. Analysis While the authors were able to point out limitations of the research, the study is very weak and leaves the audience to question the validity of the study from the start. The information studied, mobility on functional decline, is basic nursing 101. Health care providers already understand anatomy and physiology and the “use it or lose it” scenarios. The authors fail to make the connection from their research to why it is important, and what population will benefit from this information. It is best practice to mobilize patients as soon as possible, this article fails to prove new ground breaking information. The researchers did suggest that further studies are necessary and suggest that disease should be studied to see how they affect the results. This is fine, however, what difference will this make? Knowing that patients need to be mobilized and having the correct tools in place are two separate topics. Perhaps the next study should indicate the need for additional staff members that are solely dedicated to helping patients 36 GROUP CRITIQUE ambulate. The extra set of hands on the floor may prove to be more of a asset than understand the fact that patients need to move. Study Conclusions, Implications, & Recommendations Evidence Padula et al. (2009) concluded that “early and ongoing ambulation in the hallway may be an important contributor to maintaining functional status during hospitalization and to shortening LOS” (p. 330). The authors also conclude that “ambulation should be viewed as a priority and as a vital component of quality nursing care” because research indicates that “functional decline actually begins in the preadmission period” (Padula et al., 2009, p. 330). Support The conclusion includes a summary of the results or outcomes. In order to form a conclusion Burns and Grove (2011) states “the researcher uses logical reasoning, creates a meaningful whole from pieces of information obtained through data analysis and findings from previous studies, remains receptive to subtle clues in the data, and considers alternative explanations of the data” (p. 412). Risks are involved when the researcher forms the conclusion that goes outside of what the data states (Burns & Grove, 2009, p. 412). The study data must support the conclusion statements. Analysis The conclusion Padula et al. (2009) provided the audience was short and did not provide a meaningful summary of the study data. Padula et al. (2009) concluded “this research supports existing literature” but did not recommend any further research or studies concerning the mobility protocol (p. 330). Although the conclusion affirms that 37 GROUP CRITIQUE 38 ambulating in the hallway maintains functional status, the researchers do not provide the distance or how often a patient needs to ambulate to maintain functional status leaving the audience with questions (Padula et al., 2009, p. 330). The mobility protocol was not referenced in the conclusion and the researchers did not provide any other explanation to the data outcomes in the conclusion. Critique Conclusion Nursing research is vital to the nursing profession. Burns and Grove (2011) define nursing research as “a scientific process that validates and refines existing knowledge and generates new knowledge that directly and indirectly influences nursing practice” (p. 4). Nursing research helps develop a solid, evidence-based practice for the professional nurses and Burns and Grove (2009) believe that “evidence-based practice involves the use of best research evidence to support clinical decisions in practice” (p. 22). Overall, Impact of a Nurse-Driven Mobility Protocol on Functional Decline in Hospitalized Older Adults was a weak research article. Although the outcome results presented are appealing, there are many factors that may have affected the results and need to be taken into consideration when reviewing this research study. The only section found to be strong in this article was the purpose section. The authors clearly defined the purpose, their goals, and how the mobility protocol would affect those involved. All other sections in this article were found to be weak for a variety of reasons. The problem statement lacked statistical information and subject knowledge that may have drawn the audience to review the research article. The review of literature was sporadically referenced throughout the article and seven of the articles were found to be older than five years making the information less meaningful to the current knowledge. GROUP CRITIQUE The authors used the GENESIS model for the framework; however, this model was not detailed enough for the audience to fully grasp why this was chosen and the importance it had on the research. Use of a nonequivalent control group was an internal threat to validity because the control group and the comparison groups were not equal. The authors did not address co morbidities in this study leaving the audience to question if one group had more co morbidities than the other group. The participant’s previous condition could have affected the outcome and skewed the results. The sample size was small with twenty-five participants in each group. The research of the mobility protocol needs to be further studied on a larger sample size to validate the results. Also, the audience needs a better explanation of the instruments used in this study such as the reason for selecting the instruments, if they have been used in past research, and if they found them to be reliable and valid. Instruments were also found to be an internal threat to validity. Mobility of hospitalized patients is an important topic in nursing. It has been found that many complications arise from prolonged immobility. We recommend that further research be conducted and reviewed on mobility and mobility protocols in hospitalized patients. 39 GROUP CRITIQUE 40 References Burns, N., & Grove, S.(2011). Understanding nursing research: Building an evidencebased practice (5th ed), Marilyn Heights, MO: Elsevier Saunders. Padula, C., Hughes, C., & Baumhover, L. (2009). Impact of a nurse-driven mobility protocol on functional decline in hospitalized older adults. Journal of Nursing Care Quality, 24(4), 325-331.