Feasibility of a Patient-Reported Outcomes Quality of Life Instrument to Improve Care in Underserved, Ethnic Minority Cancer Patients

A better understanding of social determinants of health and psychosocial and economic sequelae of cancer for an ethnoculturally diverse population is needed. A Patient-Reported Outcomes Quality of Life (PROQOL) instrument has been developed to capture less commonly discussed patient concerns. We sought to adapt this instrument and pilot it in underserved, ethnic minority cancer patients at Maricopa Integrated Health System (MIHS) in Phoenix, AZ. Two focus groups (FG) were conducted to understand patient perspectives of an adapted Spanish version of PROQOL. A revised version based on FG data was then piloted among patients on active cancer treatment. Patients also completed a Linear Analog Scale Assessment (from 1-10) for quality of life (QOL) and its domains.

Conclusions PROQOL is feasible for systematic capture of patient concerns in underserved, ethnic minority cancer patients. A better understanding of these concerns can help design interventions to improve psychosocial outcomes for the vulnerable groups and decrease disparities in care delivery.

Background
With the Commission on Cancer's requirement for developing a process to incorporate the screening of distress into standard oncological practice, healthcare systems are now routinely administering patient-reported outcome measures (PROMs) to detect distress and to evaluate how quality of care affects cancer outcomes. [1] However, most PROMs have been developed and validated in populations with very low proportions of groups shown to be especially vulnerable to distress, like racial/ ethnic minorities or those with very low socioeconomic status. Previous studies have shown higher risks for adverse psychosocial outcomes in cancer survivors who face additional social stressors as poverty, low level of formal education, language barriers, and/or belonging to an ethnic or cultural minority. [2] As has been suggested by some cross-cultural studies, the profile for psychosocial stressors may be different in the underserved populations with diverse distributions of socioeconomic variables. [3,4] Furthermore, while there is information available about potential barriers to access that minority and underserved populations encounter, not much is known about which challenges are of greatest concern to the patient. [5,6] Aziz and Rowland have identified absence of the right measurement systems and appropriate questions to capture the ethnic and cultural differences in the experiences of cancer care as one of the major problems in addressing disparities in the delivery of high-quality cancer care to medically underserved populations. [7] The Beacon Patient-Reported Outcomes Quality of Life (PROQOL) instrument was initially developed for patients with diabetes to systematically report their quality of life at the health care visit. [8] Recently, it has been modified and piloted with a predominantly Caucasian population in a Hematology and Medical Oncology practice to better understand treatment toxicity, emotional duress from the diagnosis and barriers to improved care. [9] The instrument provides patients an opportunity to report concerns on less commonly discussed issues that may impact their health management, such as social factors and personal relationships.
The goal of this study was to address the measurement and knowledge gaps in assessing distress among underserved and ethnic minorities. We used a mixed methods (qualitative and quantitative) approach to adapt and pilot the PROQOL instrument at Maricopa Integrated Health System (MIHS) and assess its feasibility for capturing the patterns and predictors of psychosocial stressors among an underserved, economically disadvantaged and ethnic minority population.

Methods
The study was approved by the IRBs at MIHS and Mayo Clinic and all patients provided informed consent in Spanish or English. Phase I of the study consisted of two focus groups (FGs) to understand patient perspectives of an adapted Spanish version of PROQOL, using grounded theory approach. In Phase II, the PROQOL was piloted among patients receiving active cancer treatment. Refreshments were provided during the meeting held at the clinic and participants were given gift vouchers at the completion of the FGs. Both FGs were moderated by a clinical psychologist (PhD) at MIHS who was not a part of the study team and was assigned by MIHS to moderate the focus groups without any bias since not related to the study participants. Each FG lasted for approximately 60 minutes and field notes taken by NK and GR as well as transcribed audio recording of the session were used for the analysis.

Pilot Phase Patients
Inclusion criteria for the pilot phase were the same as for the FGs. A convenience sampling approach was used, and patients who were referred by their treating physician were approached to participate in the study between November 2017 and January 2018. Demographic and clinical data, including age, sex, race/ethnicity and disease characteristics, were obtained from the medical record. Median income based on residence zip code was derived from Census data and used as a proxy for patient socioeconomic status.
[10] PROQOL The PROQOL instrument system has been developed to function as a self-reported tablet-based questionnaire that can assess symptoms/quality of life (QOL). Patient responses provide treating physicians and nurses with information regarding bothersome symptoms and/or psychosocial problems that may not be addressed normally. [8] When completing the PROQOL on a tablet in the waiting room, patients select their single biggest concern from various categories such as: Personal Relationships, Emotional Health, Physical Health, Cancer Diagnosis and Treatment, Money and Care Planning. Patients also complete a 10-item Linear Analog Scale Assessment (LASA) to assess global QOL, and physical, emotional and social well-beings. LASA is a validated measure for QOL, where higher scores indicate better QOL. [11]. [12] Analysis The main aim of the focus groups was to understand the participants' viewpoint about the PROQOL instruments, including its ease of understanding, feasibility and usefulness. Using standard qualitative inductive approaches, we also analyzed the qualitative data to understand other potential psychosocial concerns. We used open and axial coding to label ideas, combine them and define categories based on overlying themes in the data.
Descriptive statistics were used to describe sociodemographic and clinical characteristics of the cohort of patients enrolled in the pilot phase of the study. A univariate analysis (Kruskal Wallis test for continuous and Chi-Square test for categorical variables) was conducted to examine the association between the most commonly identified concerns and baseline sociodemographic characteristics.

Baseline characteristics
A total of 31 subjects were consented for both FGs and a total of 12 attended the meeting. All participants in both FGs were Hispanics and most (11/12) were women.  Concerns/ QOL assessed by PROQOL in pilot phase participants Figure 1 shows results from the PROQOL instrument question, "What is your single biggest concern?". The three most common concerns endorsed by the pilot phase participants were 'Money' (32%) and 'Cancer Diagnosis, Treatment and Survivorship' (32%) and 'Physical Health' (15%). The majority of participants who endorsed "money" as a concern reported difficulty with medical bills, non-medical costs, living expenses etc. For most participants who endorsed Cancer Diagnosis, Treatment and Survivorship concerns, it was treatment plans, including chemotherapy and use of complementary therapies that caused greatest concern.
There was no difference in the distribution of concerns based on age, sex, marital status, type of cancer or ethnicity.
Median overall QOL was 8 (range 3-10), suggesting relatively high QOL. Figure 2 shows the mean LASA scores in the different domains for the FG participants.

Discussion
The increasing diversity of the U.S. population has drawn attention to the need for developing culturally relevant methods in order to integrate the physiologic, psychosocial, and economic sequelae of cancer and its treatment into routine healthcare delivery. For ethnoculturally diverse and medically underserved populations, a better understanding of their challenges is needed to design interventions to address their less commonly discussed concerns and decrease disparities in care delivery. In this mixed methods study, we were able to show that PROQOL is feasible for systematic capture of patient concerns in a medically underserved patient population with ethnic minority cancer patients. Desire for more information about the cancer diagnosis and treatment options, financial concerns and concerns regarding physical health emerged to be the most important concerns for this population. Difficulty in accrual due to patients' beliefs and attitudes is a well-known barrier for conducting such studies in an underrepresented population. [13,14] Surprisingly, our use of study staff trained to work with minority populations, communication about the study in a culturally appropriate, respectful, and easy to understand way and use of Spanish study instruments helped in accruing more than 50% of approached patients and retaining them on the study with not too much difficulty. Though seven minutes is a long time to complete the instrument, this was mostly due to the participants' discomfort in handling the tablet. A paper version was available for those not able to do it on the tablet, but all participants ended up completing the tablet version. Also, this was completed in the waiting room prior to the physician appointment, so we do feel it would be feasible without interfering much with appointment times and workflow.
Increased psychosocial distress and poor quality of life (both overall and in some specific dimensions only) have been reported extensively in cancer survivors. [15][16][17] Opportunities to screen for and improve psychosocial concerns exist throughout the spectrum of cancer care from diagnosis to end-of-life care. As the proportion of minorities in the US population continues to grow, it creates a greater imperative to understand the racial/ethnic disparities in the spectrum of psychosocial concerns that may impact cancer outcomes eventually.
Studies that have reported on quality of life or psychological problems in minorities have been done using standardized instruments to capture the outcomes. Bowen et al have shown poorer physical functioning in 'Blacks' and 'other ethnic groups' as compared to Whites. [18] Somewhat conflicting results were reported in a qualitative study of breast cancer survivors in a multi-ethnic group which revealed that the overall HRQOL was not significantly different between various ethnicities. However, the authors noted that economic barriers were particularly important for African during cancer survivorship, being African American was reported to be related to fewer cancer-related health worries. [20] Although income and socioeconomic status is closely intercalated in the web of racial/ethnic disparities, the literature on impact of income on psychosocial concerns is limited too. A study that evaluated the income disparities in quality of life among cancer survivors showed poorer quality of life in low income groups. [21] Lower household income has been related to both the treatment related financial hardship as well as adverse psychological outcomes. [22,23] Evidence suggests that the biopsychosocial burden of cancer may be particularly high for certain ethnic groups. Latinas were more likely to report poor mental, physical and social QOL when compared with non-Latina Whites and Blacks, as well as lower perceived social support, high levels of uncertainty and higher levels of depression. [19,24] Hispanic cancer survivors had the highest level of unmet supportive care needs (informational, practical, supportive and spiritual) when compared with African Americans and Whites.
[25] Data from our pilot phase also showed the difference in spectrum of concerns identified by PROQOL when administered to a predominantly Caucasian population at Mayo Clinic vs. the minority population at MIHS. [9] While 'Cancer Diagnosis, Treatment and Survivorship' was one of the two top concerns for both cohorts, 'Money' emerged as an equally important concern in the minority population as opposed to 'Physical Health' in Caucasian population.
The PROQOL also provides a platform to develop suggested actions as a template that local health care teams can adapt to reflect local community resources and their specific team's structure or method of operation. System level factors may need policy solutions targeted towards minority populations, but other potential patient-centered interventions to help address these concerns could range from increased psychosocial and tangible support with involvement of medical staff, community resources, patient navigators, and patient support groups.
There are a few limitations to our study. Because we did not have the individual level sociodemographic data, we used census data as a proxy for income. The

Ethics approval and consent to participate
The study was approved by the IRBs at MIHS and Mayo Clinic and all patients provided informed consent in Spanish or English.

Availability of data and material
The data that support the findings of this study are available from MIHS but restrictions may apply to the availability of these data, which are not publicly available. Data are however available from the authors upon reasonable request and with permission of MIHS.

Competing interests
No conflict of interest for any authors.

Funding
The project was funded by Mayo Clinic Investment for Extramural Grants Award     Figure 1 Biggest concern identified on PROQOL ISSM_COREQ_Checklist_nk.pdf