Dear Navigators, As we close out another year, I would like to thank you, our Journal of Oncology Navigation & Survivorship (JONS) reading community, for your support, encouragement, and loyalty. [ Read More ]
December 2016 VOL 7, NO 11
Quantifying the Relational Dimensions of Study Staff in a Randomized Controlled Trial Among African American Women Recommended to Receive Breast Cancer Chemotherapy
Sarah F. Gallups, RN, BSN, MPH; Mary Connolly, RN, BSN; Jacqueline Simon, BA;
Margaret Q. Rosenzweig, PhD, FNP-C, AOCN, FAAN
University of Pittsburgh School of Nursing, Pittsburgh, PA
Background: The Adherence, Communication, Treatment, and Support (ACTS) randomized controlled intervention study was a one-time psychoeducational intervention encouraging adherence of African American women to prescribed chemotherapy. Surprisingly, 98% of both the intervention and usual care groups initiated chemotherapy.
Objective: This study is a secondary aim of the ACTS intervention study to explore the influence of nonspecific factors of a largely racially concordant study team on study outcomes in the usual care group and compare the results with previously obtained debriefing questionnaire data from the ACTS intervention group.
Methods: An investigator-derived debriefing questionnaire was completed in both groups regarding nonspecific factors such as support, bond, and concern displayed by the ACTS study team or the interventionist. Descriptive statistics and independent t tests were used to analyze subscores and compare total scores of both groups.
Results: The nonspecific factors of concern for health and well-being (n = 32, M = 11.41 in usual care; n = 50, M = 11.02 in intervention) and support (M = 11.38 in usual care, M = 10.74 in intervention) were ranked highest in both groups. While the usual care group had higher mean scores overall, the mean subscores and total score (P = .4) between the groups were not significantly different.
Discussion: Regardless of intervention, all women felt a sense of support. These elements of support from study staff as well as a racially sensitive intervention may have encouraged adherence and contaminated study findings.
Conclusion: The results from this study point to the need for health disparity researchers to evaluate the impact of nonspecific factors in intervention research.
Both nationally and locally there is a disparity among African American women for initiation of recommended chemotherapy for breast cancer. As a response to this disparity, researchers at the University of Pittsburgh School of Nursing developed and tested a randomized controlled trial of a psychoeducational intervention for African American women to encourage understanding of why chemotherapy was recommended, and to address fears and concerns related to therapy. The Adherence, Communication, Treatment, and Support (ACTS) intervention was developed in an iterative fashion with the assistance of a community-based African American breast cancer survivor organization.1 At the recommendation of the community advisory group, the research staff interacting with the participants was almost entirely African American. Through the use of tailored decision-making tools and a race-matched interventionist, the ACTS intervention addressed attitudes that may impact the decision to accept chemotherapy, promoted patient communication with healthcare providers, provided tailored health information about breast cancer treatment, and offered support through the presence of an African American breast cancer survivor interventionist and tailored video messages from the African American community.1 The one-time intervention was 30 to 45 minutes in duration.
Participant enrollment concluded in September 2014. Interestingly, 97.1% (N = 69) of the usual care group and 98.5% (N = 69) of the ACTS intervention group initiated chemotherapy. This surprising result was then analyzed according to the nonspecific factors that may have influenced the results of this trial.
Nonspecific factors are a key element to consider when examining research outcomes. These factors may not be specified by theory or by the study purpose but may influence study outcomes.2-4 This is in contrast to specific factors that are “essential, theoretically derived components of the intervention” expected to produce the desired change in the dependent variable.5 Nonspecific factors in a research study, more specifically randomized controlled trials, may be a result of the participant-interventionist relationship. The formation of that relationship may be influenced by the similar background or life experiences of the participant and interventionist, and the bond that forms may affect the participant’s response to the intervention.6,7 In addition to the bond, the presence of other nonspecific factors may include the participant’s motivation to please, the participant’s outcome expectations, interventionist qualities (eg, warmth, positive regard, competence), and attention bestowed on the participant.3,5
These nonspecific factors may become even more apparent in health disparity research due to the necessity of these studies to tailor and design culturally relevant interventions. Furthermore, these nonspecific factors may come into play during the recruitment, intervention, or data collection. For example, if the disparity is race-based, an effort to create an atmosphere conducive to the enrollment of minority individuals by using race-matched recruiters could also produce the desired change in the dependent variable. Additionally, common ways for African American women to support one another and cope with a breast cancer diagnosis is through spiritual interventions such as prayer; this nonspecific factor is not part of the intervention protocol, and it may influence the participant’s receptiveness to the intervention.8-11 Nonspecific factors present in randomized controlled trials of health disparity interventions have not received adequate attention despite their ability to muddle the effects of theoretically derived intervention components. Therefore, it is imperative that researchers identify and quantify the effects of nonspecific factors in randomized controlled trials of psychoeducational interventions to accurately distinguish those effects from the efficacious components of the intervention.5
The present study is a secondary analysis from the ACTS intervention study aiming to explore the influence of nonspecific factors of a largely racially concordant study team on study outcomes in the usual care group and compare the results with previously obtained debriefing questionnaire data from the ACTS intervention group. The study protocol was approved by the University of Pittsburgh Institutional Review Board.
Participants included women recommended to receive chemotherapy between January 2011 and June 2015. Individuals were eligible to participate in the ACTS intervention study if they self-reported black race, were aged 18 years or older, had been diagnosed with invasive breast cancer at any stage, and were recommended to receive chemotherapy. Individuals were excluded if they had impaired cognition (measured by a score of less than 22 on a Mini Mental Status Exam) and were not able to understand the English language. Individual eligibility was confirmed by clinician verification that the patient was recommended to receive chemotherapy. Informed consent was obtained from all study participants. Study participants were recruited from 6 cancer centers located in western Pennsylvania and northeast Ohio.
Sociodemographic Factors: Sociodemographic factors were measured using an investigator-derived Sociodemographic Questionnaire, a 23-item form that includes variables such as age, marital status, employment status, household composition, health insurance status, and income level.
The debriefing questionnaire was the mechanism for examining the effect of nonspecific factors (ie, bond, motivation to please, attention, and interventionists’ supportive behaviors) on outcome measures, including improved adherence and improved scores on 3 proximal outcome measures (follow-through with chemotherapy, overall cancer distress, and social support). The debriefing questionnaire was completed in both the intervention and usual care groups. The questionnaires were administered in person or by telephone for the ACTS group at the completion of the study. The usual care group questionnaires were completed at a later date and by telephone.
ACTS Debriefing Questionnaire: Questions were developed based on previous research delineating supportive components of an intervention or from the interventionist.3,5 This questionnaire was investigator-derived and assessed specific components of the ACTS intervention and the overall research methodology (nonspecific effects) that were most meaningful to the ACTS participants in their decision to accept or not accept chemotherapy. The questionnaire utilized a 5-point Likert-type scale with 1 indicating “not at all” and 5 “very much” and a possible point range of 24 to 120, where a higher score indicated a greater influence of the ACTS interventionist. The 24-item questionnaire measured the amount that the suggestions of, tumor and treatment information provided by, support provided by, bond felt with, not wanting to disappoint, time and attention of, and concern for the participant’s health and well-being by the ACTS interventionist influenced their follow-through with chemotherapy, level of support felt, and distress level.
Usual Care Debriefing Questionnaire: This questionnaire was investigator-derived and assessed specific components of the overall research methodology (nonspecific effects) that were most meaningful to the usual care participants in their decision to accept or not accept chemotherapy. The questionnaire utilized a 5-point Likert-type scale with 1 indicating “not at all” and 5 “very much” with a possible point range of 15 to 75, where a higher score indicated higher influence of ACTS study staff. The 15-item questionnaire measured the amount the support of, bond felt with, not wanting to disappoint, time and attention of, and concern displayed by the ACTS study staff influenced their follow-through with chemotherapy, level of support felt, and distress level.
Data Collection and Analysis
The study data were collected in interview format through in-person or telephone interviews. The intervention group debriefing questionnaires were completed at time point 3, which corresponded to the end of chemotherapy, between August 2010 and February 2015. Since similar adherence was noted in the groups, the debriefing questionnaires in the usual care group were collected after time point 3 between November 2015 and February 2016. The data were then entered and verified in REDCap (Research Electronic Data Capture), hosted at the University of Pittsburgh.12 Data management was then conducted using SPSS version 23.13 Subscores were calculated for support, bond of shared experience, not wanting to disappoint, time and attention, and concern for health and well-being, as well as subscales for support, follow-through with chemotherapy, and distress level. Three nonspecific factors were removed from the ACTS intervention group debriefing questionnaire subscores and total score (ACTS Video influence, interventionist suggestions, tumor and treatment information) that were not included in the usual care group debriefing questionnaire. Descriptive statistics (mean, standard deviation [SD]) were used to summarize the data. Intervention and usual care group subscores and total scores for the matched questions were compared using independent t tests.
A total of 151 African American women were consented and randomized for study participation from the 6 study sites. The women were randomized to the ACTS intervention group (n = 75) or usual care group (n = 76). A total of 82 participants in both groups completed the debriefing questionnaire—50 in the ACTS intervention group and 32 in the usual care group (Figure).
The majority of participants (n = 53, 64.6%) were between the ages of 41 and 60 years, with 6.1% (n = 5) 40 years or younger and 29.3% (n = 24) older than 60 years. At baseline, 32.9% of participants had never been married, 34.1% were married or partnered, and 32.9% were not married or partnered. Regarding work, 41.5% (n = 34) of the participants had full-time employment, 46.3% (n = 38) were not working, and 12.2% (n = 10) were working part-time. These women had a range of breast cancer types and stages. Most participants had stage 1 (n = 32, 22.9%) or stage 2 (n = 67, 47.9%) breast cancer (Table 1). In the larger ACTS intervention study, there were no statistically significant differences between the study groups in demographic data or tumor type.14
The mean total score for the usual care group was 50.25 (SD 16.07). The mean subscores of the debriefing questionnaire for the usual care group ranked displayed concern for the participant’s health and well-being (M 11.41, SD 3.84), support (M 11.38, SD 3.27), and the bond of shared experience (M 11.19, SD 3.75) by the ACTS study team highest. Nonspecific factors of not wanting to disappoint (M 6.00, SD 4.37) and time and attention (M 10.28, SD 3.96) by the ACTS study team were ranked lowest. Of the larger ACTS study outcomes, support (M 17.72, SD 5.04) was rated as influenced the most by the nonspecific factors (Table 2).
Overall, the ACTS intervention group had a lower mean total score (M 47.36, SD 15.50) and lower mean subscores than the usual care group. However, like the usual care group, the ACTS intervention group rated displayed concern for the participant’s health and well-being (M 11.02, SD 3.61), support (M 10.74, SD 3.48), and the bond of shared experience (M 10.88, SD 3.40) by the ACTS interventionist highest. Likewise, not wanting to disappoint (M 5.68, SD 4.00) and time and attention (M 9.88, SD 3.67) by the ACTS interventionist were rated lowest. Additionally, the ACTS intervention group rated the study outcome of support (M 16.90, SD 5.61) as the most influenced by the nonspecific factors (Table 2). Although the usual care group had higher overall mean total and subscores, there was no significant difference between the group scores (P = .42).
Study staff may have influenced the primary results of the ACTS study through relational tasks. From these results, we find that support and concern for the patient’s health and well-being from either the ACTS interventionist or study staff may have influenced the study outcomes. Although these results cannot confirm the influence of the supportive nature of the study staff on study results, we suspect that it positively influenced the improved adherence to chemotherapy seen in both the usual care and intervention groups in the larger ACTS intervention study. These results suggest that the presence of peer support within medical oncology clinics encourages adherence. Evidence for this peer support has been shown in qualitative research from the area of patient navigation in cancer care, which found that many times both navigators and patients noted the importance of relational tasks, such as support, over the instrumental interventions, or task-oriented activities of navigators.15,16 The relational tasks include spending time with the patient, building trust and relationships, encouragement, social support, supporting autonomy, providing comfort, effective communication, and coaching.15-17 The results also identified the importance of culturally matched patient navigators. Analysis of these studies points to the importance of the patient-navigator relationship on the ability to build trust, develop rapport, provide psychosocial support, and increase the patient’s participation in care.15
While having a racially diverse healthcare system would prevent the concern of contaminating study results through the effects of racially sensitive interventions, it is unclear if the researcher’s attempts to develop a racially sensitive intervention or the combined effect of racially concordant and supportive study staff contaminated this study’s findings. The presence of racially concordant study staff may have contaminated the study findings as this was an anomaly in the local healthcare system with limited racial diversity among its staff. Although the United States is an increasingly diverse nation, ethnic and racial minorities are drastically underrepresented in the health professions. A more racially and ethnically concordant healthcare workforce would allow for a more inclusive healthcare environment for underserved populations as well as allow for greater connections and communication between patients and their providers.18,19 Realization of a more diverse healthcare workforce would create the foundation from which to conduct more inclusive and representative research on health interventions.
There were several considerations in the design of this secondary study. The difference in debriefing questionnaire completion time points between the study groups may have impacted study results, including a smaller sample size in the usual care group. However, this was unavoidable because once negative study results were appreciated and the possible etiologies were discussed, the need for a debriefing of the usual care group was made apparent. While the questionnaire was self-report and investigator initiated, it was based on well-established literature.2-5 Lastly, this study only described the debriefing questionnaire results and compared the results between groups; there was no assessment of the influence of these results on the study outcomes.
The results of this study add to the evidence for the need to further research peer support and other supportive interventions in cancer care, especially for minority and underserved groups. Furthermore, as behavioral and health science researchers are called upon to produce efficacious disparity-eliminating interventions, the need for an honest discussion regarding the presence and impact of nonspecific factors grows increasingly important.
- Rosenzweig M, Brufsky A, Rastogi P, et al. The attitudes, communication, treatment, and support intervention to reduce breast cancer treatment disparity. Oncol Nurs Forum. 2011;38:85-89.
- Budd R, Hughes I. The Dodo Bird Verdict – controversial, inevitable and important: a commentary on 30 years of meta-analyses. Clin Psychol Psychother. 2009;16:510-522.
- Chatoor I, Krupnick J. The role of non-specific factors in treatment outcome of psychotherapy studies. Eur Child Adolesc Psychiatry. 2001;10(suppl 1):19-25.
- Jensen PS, Weersing R, Hoagwood KE, et al. What is the evidence for evidence-based treatments? A hard look at our soft underbelly. Ment Health Serv Res. 2005;7:53-74.
- Donovan HS, Kwekkeboom KL, Rosenzweig MQ, et al. Nonspecific effects of psychoeducational intervention research. West J Nurs Res. 2009;31:983-998.
- Horvath AO, Symonds BD. Relation between working alliance and outcome in psychotherapy: a meta-analysis. J Couns Psychol. 1991;38:139-149.
- Martin DJ, Garske JP, Davis MK. Relation of the therapeutic alliance with outcome and other variables: a meta-analytic review. J Consult Clin Psychol. 2000;68:438-450.
- Henderson PD, Gore SV, Davis BL, et al. African American women coping with breast cancer: a qualitative analysis. Oncol Nurs Forum. 2003;30:641-647.
- Tate JD. The role of spirituality in the breast cancer experiences of African American women. J Holist Nurs. 2011;29:249-255.
- Gillum F, Griffith DM. Prayer and spiritual practices for health reasons among American adults: the role of race and ethnicity. J Relig Health. 2010;49:283-295.
- Hamilton JB, Sandelowski M. Living the golden rule: reciprocal exchanges among African Americans with cancer. Qual Health Res. 2003;13:656-674.
- Harris PA, Taylor R, Thielke R. Research electronic data capture (REDCap)—a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42:377-381.
- IBM SPSS Statistics for Windows. Version 21.0. Armonk, NY: IBM Corp; 2012.
- Jiang Y, Yee MK, Sereika SM, et al. Chemotherapy delays or dose reductions in African American women with newly diagnosed breast cancer. Paper presented at: 37th Annual Meeting of the Society of Behavioral Medicine; March 30-April 2, 2016; Washington, DC. Abstract D017.
- Jean-Pierre P, Hendren S, Fiscella K, et al. Understanding the process of patient navigation to reduce disparities in cancer care: perspectives of trained navigators from the field. J Cancer Educ. 2011;26:111-120.
- Philips S, Nonzee N, Tom L, et al. Patient navigators’ reflections on the navigator-patient relationship. J Cancer Educ. 2014;29:337-344.
- Yosha A, Carroll JK, Hendren S, et al. Patient navigation from the paired perspectives of cancer patients and navigators: a qualitative analysis. Patient Educ Couns. 2011;82:396-401.
- Cohen JJ, Gabriel BA, Terrell C. The case for diversity in the health care workforce. Health Aff (Millwood). 2002;21:90-102.
- US Department of Health & Human Services. The Rationale for Diversity in the Health Professions: A Review of the Evidence. US Department of Health & Human Services; 2006.
Treating the Whole Patient: A Patient-Centered Program Integrating Evidence-Based Complementary Therapies with Traditional Cancer Care
Ms Hoopes is an oncology certified and board certified holistic nurse and survivorship nurse navigator with Greenville Health System. She holds a bachelor of science in nursing degree from the [ Read More ]