Nursing Research


Qualitative research critique

Caring for Dying Patients Who Have Air Hunger
Qualitative Research Critique

Research is a key to developing sound scientific practice. Critical evaluation of the research is key in developing the safest and most helpful practice available to patients. Application of research to everyday practice is the goal of all research.

The phenomenon of interest was stated clearly in both the abstract and the article. The phenomenon is the experience of caring for a dying person who is experiencing air hunger. The study requires a qualitative design because the researcher is interested in the “lived experience” of caring for dying patients with air hunger. It is utilizing the phenomenological method of qualitative research. The philosophical underpinnings of this study are that breathing has symbolic as well as physiological components. The Oxford English Dictionary is used as the source of the definitions of inspiration and expiration. Foote, Sexton, & Pawlik are used as the authorities on the factors contributing to general dyspnea (Tarzian, 2000).

The purpose of the study is made very explicit in both the abstract and the text of the article. “The purpose of this study was to understand the experience of caring for a dying person who has air hunger” (Tarzian, 2000, p. 137). The projected importance is that understanding is “a step toward more consistent and informed response by health care providers to dying patients who suffer from air hunger, and to their family members who witness their distress” (Tarzian, 2000, p. 137).

The method used to collect data was an interview with the participant that was transcribed verbatim prior to analysis. This method seems adequate to address the phenomenon of interest. Van Manen developed the guidelines used for interpretation, but I did not recognize this or that anything was being followed (Tarzian, 2000).

Selection of the participants in research must be done carefully in order to obtain a sample representative of the population. The researcher does not describe the selection process of participants. She does not mention when data saturation was reached. However she does give us the demographics and all seem appropriate people to ask about dying patients (Tarzian, 2000).

The data collection does focus on the human experience utilizing a recorded interview. Pseudonyms are utilized to protect confidentiality. Saturation of data was not mentioned. Data collection was obtained through an interview that was then transcribed (Tarzian, 2000).

Themes among the participants were compared and three themes were identified. These seem consistent with what would be the experience of caring for a dying individual with air hunger (Tarzian, 2000). Having cared for dying patients in the role of a nursing assistant these seem logical and reasonable to me.

Creditability was addressed in that the nurse participants were invited to read the final thematic summary and provide feedback. Also used for creditability were interviews with two family members who witnessed their dying spouses suffer from air hunger. A panel of nurse researchers reviewed the themes that emerged and provided feedback. Auditabilty is covered well by the fact that I can follow the researcher’s reasoning and documentation. This article is fitting and has meaning if you are working where you may encounter dying patients who experience air hunger. The strategy for analysis seems compatible for the purpose and design of the study (Tarzian, 2000).

The findings are presented within the context of caring for a dying patient with air hunger. The essence of the experience is very well described using quotations and experiences of the nurse participants. The researcher’s categorizing of the data seems appropriate for what has been described by the participants. The researcher presents the data within the context of what is known about dyspnea, death, and fear of dying by suffocation (Tarzian, 2000).

The conclusion gives the reader and idea of what to expect and what to warn family members about that may develop air hunger when dying. Recommendations include a “vision of Ôdoing everything’ for a dying person with air hunger” (Tarzian, 2000, p. 143). No further research ideas are indicated, but a few can be implied. The significance of this study to nursing is implicated with the emphasis on the importance of having a preestablished plan for if air hunger develops, preparing patients and their family in advance, and responding to barriers (Tarzian, 2000).

Strengths of this study are the surfacing of the emotions behind watching someone dying. Having witnessed it three times I can identify with this study and its findings. The verbalization of the fact that difficulty breathing can be alleviated if properly treated is important. Many feel afraid to properly medicate patients for fear of killing them (Tarzian, 2000). The weakness of the study occurs because of the details that are not mentioned. Selection of participants and the lack of tracing Van Manen’s guidelines being only two examples. The risk and cost to patients if these suggestions are implemented is small. If I had to replicate the study, I would use nurses from various sectors of hospice, long-term care, emergency departments, oncology, and other settings where death rates are high. Sample size would be determined when data saturation was reached.

Understanding emotions and frustrations behind caring for dying patients with air hunger helps nurses understand how to best care for these patients. Reading and sharing emotions and actions will help to plan for the next dying patient and their families. Research and application leads to better caring.


Quantiative Research Critique

The problem statement in this article is clearly stated. “The purpose of this randomized controlled trial was to test the effectiveness of a cognitive-behavioral group intervention in reducing depressive symptoms, decreasing negative thinking, and enhancing self-esteem in young women at risk for depression” (Peden, Hall, Rayens, & Beebe, 2000, p.145). Independent variable, dependant variable and population are all easily identified. Independent variable is the cognitive-behavioral group intervention. Dependant variable is the depression symptoms, negative thinking and self-esteem. Population is college-age women who are at risk for depression (Peden et al., 2000). The way that this statement is set up it is testable.

Depression is the most common mental illness. It can impair normal life function. Negative thinking can be a good predictor for development of depression in women. Prevention of depression could be a cost effective way of dealing with what is becoming a widespread problem. Only problem is they discussed the depression rates in girls ages 14-18 and then wanted to test in the 18-24-age category. This was probably done for convenience so parental consent did not have to be considered, but this was not stated (Peden et al., 2000).

It the literature review they discussed an article that was still in press that proposed negative thinking as a mediator of the relationship between self-esteem and depressive symptoms. The interventions that should work are those interventions that work to increase self-esteem and decrease negative thinking. The abstract mentions a gap in the literature, but the article itself never specifically calls attention to the gap (Peden et al., 2000). The literature review makes it seem like this study is just testing if all of the pieces from previous studies fits together like the pieces have been suggesting. The literature review seems to be a good mix of primary and secondary articles with emphasis on the primary articles.

The hypothesis is not clearly stated but it is implied that those women in the group that received intervention would have a greater increase of self-esteem and a greater decrease of negative thinking and depressive symptoms than those in the control group, which received no treatment. It was also implied that the benefits of this treatment would persist for six months after treatment. The independent variable is the interventional treatment. The dependant variables are measures of negative thinking, self-esteem, and depressive symptoms. The way the problem statement is phrased it is a research hypothesis, not a statistical hypothesis. No research questions are asked in this article (Peden et al., 2000).

Subjects entered the study by convenience sampling, but the assignment to groups was random. This is appropriate for this type of research study. There is no mention of a power analysis to determine the needed sample size needed for this research study. These results will be generalizable to college-age women at risk for depression (Peden et al., 2000).

This is an experimental, pretest-posttest designed study. The rational used for that design is that the control and experimental group need to be as similar at the beginning as possible so that the change is attributable to the intervention (Peden et al., 2000). The design is logical for the type of study being done.

The threats to internal and external validity are not very well discussed. Instrumentation is well documented with validity and reliability scores mentioned. The article says that the data on self-esteem, depressive symptoms, and negative thoughts were collected by self-report questionnaires. No mention of where the questionnaires were completed or who actually completed them. No mentions of testing effects, history, mortality, or selection bias were addressed in the sample or design section. No external validity problems were mentioned. The self-report questionnaires are used for all research participants. No differences in administration were mentioned. No mentions were made as to when in the semester were these test administer. As all college students know, depression can be linked to what part of the semester the testing is administered. The article mentions in the data analysis that mortality occur, but a mention of what rate or mortality was never mentioned. This was not mentioned at all in method or sample section. Hawthorne effects were discussed in the discussion. Both the experimental group and the control group showed beneficial effects, but the control group showed only a slight change compared to that in the experimental group (Peden et al., 2000).

The rights of research participants were protected by preapproval by an institutional review board prior to beginning the study. Confidentiality is protected by not mentioning the participants in any identifiable way. No mention of informed consent is made (Peden et al., 2000).

Instrumentation used were only questionnaires of a Likert type questions. The Beck Depression Inventory (BDI) shows good validity measuring clinical depression and good test-retest scores. The Center for Epidemiologic Studies-Depression scale (CES-D) has been shown reliable and valid in many studies. The Crandell Cognitions Inventory (CCI) used to measure negative thoughts has been supported by construct validity. The Rosenberg Self-esteem scale (RSE) used to measure self-worth or acceptance has been shown to have excellent internal consistency and construct validity (Peden et al., 2000). The scores and reports make these tests the logical choice to use.

The level of measurement used was ordinal, but the Likert scale could be argued as an interval measurement in some cases. The methods used were chi-squares and analysis of covariance (Peden et al., 2000). Chi-squares are appropriate tests of differences between means because they are nonparametric for ordinal numbers. The analysis of covariance is debatable to be used as parametric data. It would only be acceptable if the Likert scale were used as interval data (LoGbiondo-Wood & Haber, 1998).

The hypothesis tested was supported. It seemed as though at risk college age women benefited from a cognitive-behavioral group intervention as a preventive for depressive symptoms and negative thoughts. Both groups show an improvement, but it was larger in the experimental group. The data also showed improvement in self-esteem. Tables in the text do a good job of presenting the CES-D scores and the BDI scores, but the RSE and CCI are not presented. The inferred statistics are used appropriately to support the hypothesis. Emphasis is placed on the impact of depression on the public and private health systems and how much can be saved by prevention rather than to wait for a problem to develop (Peden et al., 2000).

This would be most applicable to women at risk for depression. Most of the study appears valid and as the risk of harm to a patient is very low, I would see about trying to apply it, as it was needed.

Reference:

LoBiondo-Wood, G., & Haber, J. (1998). Nursing research: methods, critical appraisal, and utilization, (4th ed.). St. Louis: Mosby.

Peden, A. R., Hall, L. A., Rayens, M. K., & Beebe, L. L. (2000). Reducing negative thinking and depressive symptoms in college women. Journal of nursing scholarship, 32 (2).