Critical Appraisal: Living with Heart Disease After Angioplasty: A Qualitative Study of Patients Who Have Been Successful or Unsuccessful in Multiple Behavior Change

Peterson et al. conducted a qualitative research study regarding the success or failures of patients in making lifestyle changes following angioplasty. The researchers’ qualifications indicated clear knowledge of the field of study, with researchers having diverse areas of expertise in complementary and integrative medicine, health and behavioral studies, sociomedical sciences, psychosocial and health services and surgery (Peterson 105). The study, published in the National Institute of Health was clear, concise and generally well-written overall (Peterson 105-115). The study followed a logical pattern of progression and was easy to read. There were no grammatical or other fundamental errors found. The title of the research study gave the reader a clear expectation of the subject matter of the study, as did the abstract, which provided a brief but comprehensive overview of the study, while still touching on all of the salient points of the study.

The phenomena of interest in the study are clearly identified by the researchers, both in the title of the study and again throughout. The subjects of the study were a demographically and culturally diverse group of patients post-angioplasty (Peterson et al. 105-106). The researchers took care to adequately represent both women and men, successful and unsuccessful patients (criteria for this label is defined further in this critique), and still represent three major minority groups (as identified by the researchers). The study group was clearly identified and classified as unique, as this group of patients had participated in a “parent study” named the “Healthy Behavior Trial” where the patients were educated about risk-factors and provided information on behavior modification (Peterson et al. 106-107).

The purpose/significance of the study, however, is not clearly identified in the objective of the study (Peterson et al. 106). The stated objective of the study is “to document the values, attitudes, and beliefs among a diverse group of patients post-angioplasty” (Peterson 105). While the objective reads that the study was conducted to document factors that influence behavior change in the group, it is not until the conclusion that the significance of the study is disclosed -- the information was intended to determine what types of lifestyle interventions would be most appropriate to promote positive behavior changes in these patients (Peterson et al. 106). The researchers did, however, justify the use of a qualitative approach throughout the study, as well as the semi-structured interview and grounded theory methodology used in the study.

A comprehensive literature review was conducted by the researchers and is adequately presented in the study (Peterson 107; 113-114). While the review did provide an objective account regarding research conducted on coronary artery disease, a gap in the literature was noted by the researchers regarding studies that involved patients that underwent percutaneous transluminal coronary angioplasty (PTCA) (Peterson et al. 107). It was noted that the majority of research conducted in the field that was subsequently reviewed by the researchers was related to post-myocardial infarction patients and not post-PTCA patients (Peterson 107). A second, less clear purpose for the study was also stated in the conclusion, where the researchers stated that the study was conducted to “address [the] gap in the literature and advance the science of what is known” (Peterson 107). However, it is unclear from the conclusion whether this objective was truly intended to become a purpose of the study, or if the contributions to the literature would serve as an added bonus. Contrary to the premise of grounded theory (one of the stated methodologies), the literature review was conducted prior to data collection (Peterson 107). The researchers also failed to follow the principles of grounded theory in other areas, and that failure will be further discussed later in this critique.

Early in the study, the theoretical framework for the study was clearly identified as grounded theory (Peterson 105). In identifying the core variables (as defined by Polit & Beck), the researchers may address factors that contribute to unsuccessful behavioral change in patients (498). As stated in the study, the purpose of the study was to collect information in order to determine what types of lifestyle interventions would be most appropriate to promote behavior change in post-PTCA patients (Peterson 107). As Peterson et al. further describe, the researchers sought to offer insight into behaviors and provide guidelines as to which treatment was most appropriate, as such, the use of grounded theory was justified (Peterson 109).

Quantitative data in the form of baseline demographic information was collected for participants during their participation in the Healthy Behavior Trial parent study and was used again in this study (Peterson 109). Researchers did not provide information on the collection methods used for this information, nor did they justify the collection approach, but instead relied on published methods for collection of the quantitative data (Peterson 109). However, information was published in this study regarding the comparison of successful and unsuccessful patients and the testing administered (Peterson 109). The researchers asked patients a series of open-ended questions during in-depth, semi-structured interviews to collect data (Peterson 107-108; 109). Researchers reported that the initial interviews were exploratory and included a discussion with the patients regarding their perceptions of the causes, symptoms, and treatment for cardiac disease, as well as information regarding their individual experiences and attitude toward the disease (Peterson 108). The data was collected appropriately and researchers justified this data collection approach within the study, as the information was to be used to confirm, refine or refute other data collected in the study (Peterson 108).

The sample for this study was unique, in that the subjects were identified for this study because they were originally part of the Healthy Behavior Trial parent study, yielding the most appropriate (who can best supply relevant information) group for participation (Peterson 106; Polit & Beck 527). This Healthy Behavior Trial parent study was comprised of 660 post-PTCA patients who were educated about risk-factors and provided information on behavior modification (Peterson et al. 106-107). The sample size for this study was clearly identified, and 61 patients comprised the study sample (Peterson 108). The purposive sampling of participants (selection of patients that would most benefit the study) was combined with maximum variation sample techniques (intentional selection of patients to create diversity and incorporate a wide range of individual perspectives into the study) to ensure that the researchers would be able to identify beliefs specific to any one group, while at the same time allowing researchers to identify beliefs common to all groups (Peterson 108; Polit & Beck 517). Maximum variation also preserved the integrity of the sample by the deliberate selection of participants, ensuring that a diverse group of patients was represented in the sample – male, female, African-American, Latino and Caucasian (Peterson 108). The sample size exceeded the typical sample size in grounded research of 20-30 people, however this may have been appropriate given the purposive sampling also conducted (Polit & Beck 523). Data saturation (according to Polit & Beck, the point at which no additional information is obtained and redundancy is achieved) was achieved for the group (Peterson 108; Polit & Beck 521). Assuming true data saturation was achieved, the sample size was adequate (yielded sufficient and quality data) (Polit & Beck 527).

The anonymity of the patients participating in the Healthy Behavior Trial parent study may have been compromised, as it is unclear from the report how these subjects were first identified in order for the researchers to be able to contact them and extend invitations to them to participate in the new study (Peterson 108). Although the patients were reportedly provided with informed written consent in the Healthy Behavior Trial parent study, it is unclear whether that consent covered the release of information for this purpose. However, that may be an issue for the researchers in the Healthy Behavior Trial parent study and not this one. Further, only patients who completed the two-year follow-up in the Healthy Behavior Trial parent study were eligible for this study (Peterson 108). Patients deemed successful (changing two or more of 12 targeted health behaviors and maintaining the change for 12 months) or unsuccessful (unable to maintain change of two or more of 12 targeted health behaviors for 12 months) in the Healthy Behavior Trial parent study were also identified, and an equal number of those patients were selected for this study (Peterson 108).

According to the researchers, Participants in the Healthy Behavior Trial parent study were provided informed written consent in accordance with the institutional review board procedures (Peterson 107). In this study, each participant was also provided with an informed written consent prior to any structured interviews (Peterson 108). There was no mention in the study of patient confidentiality or preservation of autonomy, although both are critical for participation in research studies (Polit & Bick 557). There were no other special ethical considerations for this study.

The data collection strategies were described in detail in the study (Peterson 105-115). Using a “sample angioplasty qualitative interview question” sheet as a guide researched conducted semi-structured interviews with participants to collect data (Peterson 105). As described in the study, the interview questions revised following a “pilot-test” to guarantee quality (108). The interviews were exploratory in nature and, as suggested by Polit & Beck, the participants were encouraged to provide information using a series of probes designed to foster discussion to the point of data saturation (Peterson 108; Polit & Beck 537). Data saturation was achieved (Peterson 108). Each of the interviews was all conducted by a trained interviewer, where training consisted of a three comprehensive half-day training session with a qualified expert in the field (Peterson 109). The interview training process was described at length in the study and sufficient to meet the burden of inquiry (Peterson 109).

As outlined in the study, quantitative and qualitative data analysis was conducted (Peterson 109). The qualitative data was reportedly analyzed using grounded theory methods (Peterson 109). The collection process was outlined in sufficient detail in the study, and verification strategies were also provided by the researchers. The semi-structured interviews were “recorded by audiotape and then transcribed and analyzed line-by-line using the software package Ethnograph version 5.0” to facilitate open-coding methods (Peterson 109). The researchers were justified in their selection of Ethnograph 5.0, as the software provided a complete and accurate transcription of the interviews, a critical part of the semi-structured interview process (Polit & Beck 557). Use of the software was also justified, in that the verbatim transcripts generated by the program reduced the three categories of transcription errors – deliberate alternations of the data, accidental alternations of the data and unavoidable alternations of the data (Polit & Beck 557). There was reportedly no modification of the data by the researchers during the transcription process.

The study relies on the Strauss and Corbin approach to grounded theory analysis (Peterson 109). Strauss and Corbin (ctd. In Peterson) offer that open-coding provides a mechanism for data to be broken down into parts, then concepts are identified and their properties and dimensions are delineated (Polit & Beck 573). The researchers further justified the selection of the software in this area, as a key purpose of the study was to allow researchers to identify beliefs common to all groups (Peterson 109). As stated in the study, “four independent, trained corroborators reviewed the transcripts and generated the concepts, categories and themes” providing adequate verification of the data (110). The desired data saturation was also reportedly achieved with these interviews (Peterson 109).

Rigour for the study was also established using the four quality criteria established by Guba and Lincoln, cited by Polit & Beck -- credibility, dependability, transferability, and goodness (584). The term “credibility” refers to the primary goal of qualitative research and established by (i) carrying out the study in a way that enhances the believability of the findings; and, (ii) taking steps to demonstrate credibility in research reports (Polit & Beck 585). The selection of the study group based on their participation in the Healthy Behavior Trial parent study added credibility to the identification of post-PTCA patients (Peterson 107). The utilization of a uniform interview guide for all patients increased the believability of the findings, as all program participants were asked the same series of questions (Peterson 108). The researchers’ utilization of expert transcribers also ensures that the study was conducted with rigor (Polit & Beck 543; Peterson 109). The researchers took significant steps to demonstrate credibility in their research report by presenting the information in a clear, detailed and logical fashion. One significant issue with the credibility of the study, however, was the introduction by researchers of “themes” that emerged during the research. Themes are also mentioned by the researchers throughout the study and relied upon in the conclusion. According to Polit & Beck, “grounded theory study should not present themes, because grounded theory analysis does not yield themes” (577). While according to Polit & Beck it is common to “muddle traditions,” this error in the data should be noted (Polit & Beck 577).

The term “dependability” refers to the stability of data over time and conditions (Polit & Beck 585). The study followed patients who were three years post-PTCA, and who had been recommended “a novel approach to behavior change” (Peterson 107). While the study appears to accurately capture “the values, attitudes and believe that influence behavior change,” it is unclear whether these findings would remain true in the long term (Peterson 107). When asking the dependability question of whether the findings would be repeated if the study were replicated in the same context, the answer is presumably yes (Polit & Bick 585). However, a true test of the information would be to conduct the same study in another three years, and possibly even after a longer period of time, to verify whether the initial findings by the researchers were accurate. The researchers do not make such a recommendation in this study.

The term “confirmability’ refers to the objectivity of the study and whether the data would be confirmed by independent evaluators (Polit & Bick 585). As stated in the study, triangulation techniques were used to interpret the data (Peterson 110). Triangulation provides that a researcher use multiple methods of reviewing the data to examine the data and create additional insight into the subject (Polit & Bick 585). The qualitative data from this study were compared with the quantitative data from the Healthy Behavior Trial parent study, confirming its validity (Peterson 110). Interviewers were trained using standardized training techniques and conducted semi-structured interviews using the same series of questions (Peterson 109). Further, four, independent, analysts reviewed the transcripts and generated the concepts, categories, and themes, again achieving the criterion (Peterson 110). The combination of these factors created trustworthiness (Peterson 110).

The term “transferability” refers to the ability of findings to be applicable to another group or setting (Polit & Beck 585). The identification of behaviors causing patients to either be successful or unsuccessful in making behavior changes could easily be transferable to other settings, where the reasons for the failure to make changes could be similarly applied. For example, in this study, a patient’s physical limitations were found to be a reason why patients did not engage in cardiac rehabilitation (Peterson 113). The results of this study are not limited in context, and the same findings could easily be applied to other post-operative patients who need to be exercise proper nutrition and physical activity in order to participate in rehabilitation activities. The same holds true for lessons learned from patients who were successful in implementing behavior change. A patient who viewed a health crisis as a “life-threatening event” could be suffering from any combination of ailments, making the findings easily transferrable (Peterson 113).

The term “authenticity” refers to the fair and faithful presentation of information in a study (Polit & Beck 585). The researchers collected information from participants during in-depth semi-structured interviews regarding their perceptions of cardiovascular disease, as well as information regarding their individual experiences with the disease (Peterson 108). The researchers included excerpts from these participant interviews, which helped the reader understand the participants’ feelings which motivated the behaviors of post-PTCA patients (Peterson 111-113). It is one thing to present data in a study regarding a patient’s complaint of physical limitations, but far more authentic to read that the patient actually complains of “run[ning] out of gas” (Peterson 113).

The findings of the research study were clearly and logically presented using a description of the phenomena and review of themes common to all patients (Peterson 113). The findings were discussed, at length, and divided into lessons learned both from successful patients and unsuccessful patients (Peterson 114). The findings were also presented with additional literature review, and used to distinguish and further validate the findings in this study (Peterson 114).

While the study was reported as being significant to identifying the underlying reasons for the success or failure of patients to modify behaviors post-PTCA, one significant limitation of the study was noted by the researchers (Peterson 115). The participants in this study previously participated in another intervention study (Peterson 115). While this commonality added credibility to the group selected, it could adversely impact the transferability of these findings to other post-PTCA patients who had not previously been exposed to such counseling (Peterson 110). Another recommendation for further study, in addition to conducting the study at another three-year interval, should be the study of post-PTCA patients who did not participant in the Healthy Behavior Trial parent study. This recommendation, however, was not included in the study. The study did make recommendation as to how the findings of this report may be developed and offered that health care providers can use information regarding patients’ values, attitudes and beliefs of post-PTCA patients to develop individualized behavior plans, instead of implementing a standardized plan for all patients (Peterson 115). This application into real-life treatment options makes the findings of the study more valuable.

Lastly, the reference an accurate and comprehensive list (51 in total) of all of the materials used in the study was appended to the study. It is easy to obtain additional information on the materials used in the study.

References

Polit, D. F., & Beck, C. T. (2011). Nursing Research: Generating and Assessing Evidence for Nursing Practice. (9th ed.). Philadelphia: Lippincott.

Peterson, J. C., Allegrante, J. P., Pirraglia, P. A., Robbins, L., Lane, K. P., Boschert, K. A., et al. (2010). Living With Heart Disease After Angioplasty: A Qualitative Study Of Patients Who Have Been Successful Or Unsuccessful In Multiple Behavior Change. Heart & Lung: The Journal of Acute and Critical Care, 39(2), 105-115.