Background
Screening for psychosocial distress in patients with cancer has become a well-established component of integrative oncologic care, as timely identification of distress and appropriate referrals have been shown to improve well-being, quality of life, and potentially even survival.1–3 As such, distress screening is recommended by NCCN, the American College of Surgeon’s Commission on Cancer, and the National Academy of Medicine.4–6 Distress screening is particularly important in patients at elevated risk for distress, including racial and ethnic minorities, women, younger patients, and those with preexisting psychological and/or substance use disorders.7–12 These demographic differences are relevant, because undiagnosed and untreated distress is not only troublesome in itself, due to direct impact on quality of life, but also contributes to cancer treatment adherence and mortality.13–15
Unfortunately, distress screening and management are laborious with significant heterogeneity in implementation across hospitals and healthcare systems.16–18 The importance and efficacy of various screening tools have been debated, including the type of instrument used for screening and which personnel should be tasked with administering screening, interpreting screening, and triaging and placing referrals for distressed patients.19–21 Parts of this workflow can fall to oncology social workers, care coordinators, clinic staff, physicians, patients, or a combination thereof. One method to simplify and reduce the burden of the distress workflow is to electronically administer distress surveys that automatically prompt follow-up of patients who screen positive for significant distress, effectively eliminating all effort needed from the clinical care team in the distress screening process until a distressed patient is identified. To date, few studies have reported on distress screening outcomes using this novel method of distress screening, and none have evaluated its implications.22
Electronic tools embedded within electronic health record (EHR) systems have been effectively used in noncancer settings, such as screening for alcohol use disorder, depression, and adolescent health risk behaviors.23–25 However, patients in rural areas, those with lower educational attainment and income, and those with advanced age report using the internet less than their counterparts.26 The intersection of heterogenous digital literacy among patents with EHR-based screening tools is largely unstudied, but there is cause for concern given these findings. One of the many unfortunate lessons learned from the COVID-19 pandemic is that healthcare disparities in virtually every realm have been amplified; with the prompt transition to telehealth, those without access to the internet were identified as a particularly high-risk group for being left behind.27
We therefore felt it imperative to investigate whether use of electronic screening for cancer-related distress may contribute to barriers in access to quality cancer care. We focused on deployment of EHR-based screening tools for cancer-related distress, which have yet to be studied and for which inequities related to access and use of electronic screening tools could have significant impacts on patient well-being and coordination of care. We hypothesized that patient access to and engagement with EHR-based screening for psychosocial distress would positively correlate with factors typically associated with digital literacy, such as younger age and higher socioeconomic status.
Methods
Electronic Distress Screening Protocol
Starting in June 2014, Oregon Health & Science University (OHSU) began an institution-wide oncology distress screening protocol. Due to the minimization of work provided, distress screening was primarily administered via an online EHR-based patient portal (Portal, MyChart). The screening instrument consisted of the Patient Health Questionnaire 4 (PHQ-4), to assess anxiety and depression symptoms, and 4 Likert-type questions scored from 0 to 10 to assess insurance/financial, strength, memory, and family-related distress (0 = no distress, 10 = high distress), as previously described.22 Patients were sent a Portal-based prompt to participate in electronic screening prior to their second in-person oncology clinic visit. Screening surveys were trigged automatically by the EHR Portal upon the scheduling of the first “established patient” encounter with an oncology provider, and therefore required no staff time for screenings to be assigned or sent to patients. Response thresholds for screening were set, and surveys were automatically scored by the EHR such that responses indicating levels of distress above set thresholds automatically flagged and routed the patient’s chart for follow-up by an oncology social worker. Within several business days, social workers then performed further evaluation and referral to support services as clinically indicated. In this way, oncology staff effort was only used to respond to patients meeting our institutional criteria for high distress, rather than to perform or interpret initial screening surveys. Follow-up with cancer social workers at our institution was readily available if distress was separately identified by an oncology provider; however, this workflow was not protocolized and patients were not screened through supplemental means.
Patients without an active Portal were routinely given an activation code and prompted to activate their account following their first visit with an oncology provider, but were not otherwise prompted to participate in distress screening. Notably, patients with Medicare insurance receiving chemotherapy were offered paper-based screening as part of the Centers for Medicare & Medicaid Services Oncology Care Model, but were excluded from the study population.
Population Identification
Patients seen at OHSU from 2014 through 2019 were identified through our institutional cancer registry, which is formatted according to North American Association of Central Cancer Registries (NAACCR) standards. For identified patients, all in-person visits at a qualifying oncology clinic were extracted from the electronic medical record system (Epic Caboodle data warehouse) using the SAP BusinessObjects Business Intelligence tool and Research Data Warehouse. Portal status was defined as “active” or “inactive” for each visit. Patients eligible for distress screening, agnostic of Portal activation status, were identified by the number of in-person oncology clinic visits, with ≥2 visits triggering Portal-based screening. Patients were further subdivided based on an active Portal status at the time of qualifying visits, such that patients who were offered an electronic distress survey at any point during their care were separated from those who were not.
Data Collection and Handling
Data from OHSU’s NAACCR-formatted cancer registry were used for analysis. Additionally, patient zip code was used to estimate patient income and educational attainment using 5-year estimates over the 2014–2019 period, based on United States Census data for patient zip code of residence at the time of first contact.28,29 Travel distance in kilometers was calculated from the patient zip code of residence to OHSU, as previously published.30 Additionally, preferred patient language and the number of prior encounters at OHSU prior to the date of first cancer-related contact were extracted from the EHR.
Statistical Analysis
Clinicopathologic characteristics were tabulated and evaluated with Fischer’s exact test and Student t testing, as appropriate. Odds of being offered electronic distress screening, which required an active Portal, were analyzed via univariable and multivariable logistic regression, as were odds of survey response in patients offered screening. Multivariable analysis was performed by single-backward elimination using the likelihood ratio test; all predictors with univariable P<.2 were included in initial multivariable models. Variables were progressively eliminated until further elimination would result in a statistically significant decrease in model fit with P<.05. For patients with missing data for continuous variables, the mean value of the cohort for the variable was imputed, as <1% of data for such variables was missing and multiple imputation was not felt to be warranted.31,32 Additionally, excluding patients with missing continuous variable data did not meaningfully change results. For categorical variables, patients with missing data were analyzed as “unknown/other.” As with continuous variables, there were few patients with missing categorical data, constituting at most 5% of specific datapoints, with the exception of insurance payer, which was unavailable in approximately 18% of patients. Outliers for continuous variables were defined as values deviating from the median >1.5 times the interquartile range, and analyses were performed both including and excluding patients with outlier values for age, estimated educational attainment, and estimated household income, with no difference in the significance of results; analyses using the complete cohort are shown (Tables 1, 2, and 3). Adjustments for multiple comparisons were not performed due to the large sample size and given that there were 2 outcomes of interest (being offered screening and responding to a screening survey). All statistical analysis was performed using SPSS Statistics, version 26 (IBM Corp). This study was approved by the OHSU Institutional Review Board and was conducted in accordance with the principles set forth in the Helsinki declaration. Data are available upon reasonable request to the corresponding author.
Results
Clinicopathologic Characteristics
A total of 17,982 patients were seen for cancer treatment or surveillance at an oncology clinic during the study period (Table 1). Most (n=10,279; 57%) had active Portal status during the study period and were therefore offered electronic distress screening. Patients offered screening differed significantly from patients not offered screening in almost all evaluated characteristics, including age, gender, race, ethnicity, primary language, disease site, insurance payer, extent of cancer care at OHSU, estimated yearly household income, estimated likelihood of having a bachelor’s degree or higher, and travel distance to our institution’s main campus.
Clinicopathologic Characteristics of Patients Eligible for Distress Screening
Odds of Being Offered Electronic Distress Screening
On univariable analysis, earlier year of first contact, older age, male gender, non-White race except Asian, Hispanic ethnicity, non-English primary language, not having private insurance, lower estimated income, lower likelihood of bachelor’s education or higher, farther distance from OHSU, less cancer-related care at OHSU, and fewer prior encounters at OHSU were associated with lower odds of being offered electronic distress screening (supplemental eTable 1, available with this article at JNCCN.org). These variables were also independently associated with lower odds of being offered distress screening on multivariable analysis (Table 2), with the exception of longer distance to OHSU, which was independently associated with higher odds of being offered distress screening after adjustment.
Multivariable Odds of Being Offered Electronic Distress Screening
Odds of Distress Screening Participation
Of the 10,279 patients offered electronic distress screening, 3,758 (36.6%) responded. On univariable analysis, factors associated with screening participation were similar to those associated with Portal activation and being offered screening (supplemental eTable 2). Generally speaking, non-White race, Hispanic ethnicity, non-English primary language, non-private insurance payers (with the exception of supplemented Medicare), farther distance from OHSU, lower estimated income and educational attainment, and lesser extents of treatment at OHSU than diagnosis and all first-course treatment were associated with lower odds of electronic screening survey response. Conversely, older age groups and a higher number of prior encounters at our institution prior to diagnosis were generally associated with higher odds of survey response. Notably, unlike being offered electronic distress screening, a later year of first contact was associated with lower odds of survey response.
On multivariable analysis, later year of first contact, Black race, Hispanic ethnicity, non-English primary language, lower proportions of first-course treatment at OHSU, and Veteran’s Affairs and Medicaid insurance coverage were independently associated with less likelihood of responding to offered surveys (Table 3). In contrast, patients with supplemented Medicare insurance (relative to private insurance) and higher estimated educational attainment were more likely to respond to offered surveys.
Multivariable Odds of Electronic Distress Screening Participation Among Patients Offered Screening
Discussion
We originally hypothesized that EHR-based distress screening may introduce disparities in this aspect of comprehensive oncologic care. Indeed, we show that patients who are older, male, non-White, Hispanic, and non-English speaking are significantly less likely to be screened in a system that relies on patient participation in an electronic healthcare portal, as are individuals with nonprivate insurance payers and those who receive any other part of their medical care at outside healthcare systems. In contrast, higher educational attainment and median household income increase the likelihood of EHR Portal activation and therefore being offered EHR-based distress screening. Furthermore, in patients offered distress screening, many of these factors were also associated with the likelihood of screening participation. We suspect that most, if not all, of these disparities can be explained by demographic and socioeconomic differences in digital literacy and engagement between patients with and without active Portals, and between survey responders and nonresponders. Together, these data suggest that EHR-based methods of distress screening in patients with cancer, although efficient, may introduce disparities in offered distress screening as well as response rates without a supplemental means of capturing unscreened or nonresponding patients. These findings are particularly relevant, as many of the underscreened groups have been identified as being at higher risk for psychosocial distress, such as the underinsured and those of lower socioeconomic status.22,33,34
Prior studies of distress screening have largely used paper surveys administered in clinic; although this process captures nearly all patients offered screening, it is laborious in the context of the multiple other tasks required of clinical staff, and is not as feasible with telemedicine.7–18 Our program is highly efficient in that no ongoing effort is required to administer or input survey results once the system is operational, and providers only become aware of a patient’s distress level when they meet predefined “high-risk” thresholds, which are adjustable. This automated process saves an estimated 3 to 4 minutes of staff time per patient to administer the survey and enter responses into the chart, plus additional time saved by the automatic screening, scoring, and flagging of distressed patients for further evaluation. Unfortunately, not all patients can or will engage with the electronic platform, as shown by the present results. Additionally, it is unclear whether the preclinic electronic format is beneficial or harmful in eliciting accurate responses compared with in-person clinic screening; these questions are important but remain entirely unstudied. It is important to recognize that due to societal stigma around discussion of mental health issues and other domains of psychosocial distress (eg, financial, interpersonal), the setting, modality, and timing of screening may have a measurable impact on responses.
Although no prior studies have specifically investigated electronic distress screening workflows, 2 prior studies have investigated patterns of use of web-based self-management distress tools for cancer-related distress. Berry et al35 reported their experience with the Electronic Self Report Assessment – Cancer (ESRA-C), showing that individuals who voluntarily used their intervention tool had decreased distress levels, but that level of education and working status were associated with voluntary use. In contrast, in their Breast cancer eHealth (BREATH) study, van den Berg et al36 did not find any sociodemographic differences among breast cancer survivors demonstrating attrition with use of a web-based self-management tool. The applicability of these studies to the current question is unclear given significant population and intervention differences from the present study, as these studies involved neither screening nor EHR-based tools.
Younger, more educated, and higher-income individuals in the United States access the internet on a much more frequent basis than their counterparts, according to national survey data.37 For instance, 99% of individuals aged 18 to 29 years in the United States reported using the internet in 2021, compared with 75% of those aged ≥65 years. Lower-income homebound older adults have been identified as a particular subset at risk for being lost in the “digital divide.” This is not to say that internet access or use is the sole determinant of engaging in an electronic healthcare system, but it is a clear first step. Regarding screening participation, in our cohort only 37% of patients offered screening chose to participate, and there is clearly much room to improve, both in access to and engagement with digital healthcare delivery systems. On this note, we highlight the finding that the adjusted odds of screening response among patients offered electronic distress screening decreased yearly during the study period, whereas the proportion of patients offered screening increased yearly. The reasons for this finding are unclear, but this constellation of findings may reflect an increasing number of patients successfully encouraged by OHSU’s institution-wide initiatives to activate their Portal account, but who did not subsequently use them to engage with their healthcare team. We were unable to measure the degree or frequency of Portal use in our cohort with active Portals, and it may have revealed a growing population of patients who fit this category. As such, future studies of efforts to improve consistent engagement of patients with EHR Portals are needed, as achieving an active Portal status alone may be insufficient to ensure subsequent Portal-based engagement.
Telehealth has been touted by many as a way to break down disparities in access to healthcare, and “meaningful use” of the EHR has been recommended over the past decade.38 Many of the disparities in accessing and using EHR-deployed electronic screening are consistent with studies both prior to and after the rapid adoption of telehealth with the COVID-19 pandemic. However, these data highlight the importance for individual medical centers to monitor their center-specific trends, as it can allow for identification of subpopulations at risk for not receiving electronic screening, allowing for targeted deployment of limited personnel resources to these groups. This could include the use of paper-based methods or deployment of additional resources to help facilitate use of the Portal, as described recently.39,40 Additionally, patients with cancer, especially the elderly, have been found to be receptive to electronic or mobile communication with their providers and may particularly stand to benefit from web-based screening and intervention.41–44 Although EHR-based distress screening suffers from limitations, its clear advantages in efficiency mandate its revision and continued iteration to become a more inclusive aspect of cancer care.
Our study is limited by several factors, chiefly by its retrospective nature. Although EHR data are often unreliable, our study primarily relied on variables collected by an accredited NAACCR-formatted cancer registry, which has a set data dictionary and standards for data reporting and quality checking. Despite this, we were unable to obtain patient-specific information on educational attainment or income, and therefore it is possible that the observed associations of race and ethnicity with receiving distress screening may merely be reflecting racial disparities in these variables within the same zip code, rather than a true independent association of race with electronic patient Portal use. Notably, 95% of Hispanic individuals use the internet, compared with 93% of White and 91% of Black individuals.37 Neither Berry et al35 nor van den Berg et al36 identified race or ethnicity as a factor in voluntary use of their web-based tools. The association of longer travel distance with Portal activation may similarly identify patients who had the financial resources to travel longer distances to our center and were more likely to be active Portal users. Additionally, although there were significant disparities in patients offered screening in the present study, the persistent disparities in completion of screening suggest that mere access to the survey is not the only barrier. Our study is unable to assess whether this relates to digital literacy, usability of the interface, trust in the healthcare system at large, or other factors. Additional prospective study of the present findings is therefore warranted.
Conclusions
Although electronic Portal-based distress screening has the potential to dramatically improve screening efficiency for patients with cancer, thereby permitting screening of larger patient volumes, this approach comes at the cost of underscreening patients who are unwilling or unable to engage in electronic modalities used for screening. These patients are generally older, non-White, and of lower socioeconomic status than those who are engaged in electronic healthcare Portals, and may be more likely to experience significant psychosocial distress.22,33,34 We therefore recommend continued innovations to improve Portal participation at institutions utilizing EHR-based distress screening, as well as supplementation with paper-based or other in-clinic distress screening to achieve efficient yet equal-opportunity psychosocial care for patients with cancer. Finally, we also recommend investigations into whether in-person versus electronic screening yields different results in screened patients, which is entirely unstudied but may have important clinical ramifications.
References
- 1.↑
Bultz BD, Carlson LE. Emotional distress: the sixth vital sign in cancer care. J Clin Oncol 2005;23:6440–6441.
- 2.↑
Faller H, Schuler M, Richard M, et al. Effects of psycho-oncologic interventions on emotional distress and quality of life in adult patients with cancer: systematic review and meta-analysis. J Clin Oncol 2013;31: 782–793.
- 3.↑
Stagl JM, Lechner SC, Carver CS, et al. A randomized controlled trial of cognitive-behavioral stress management in breast cancer: survival and recurrence at 11-year follow-up. Breast Cancer Res Treat 2015;154: 319–328.
- 5.
Adler NE, Page AEK, eds. Cancer Care for the Whole Patient: Meeting Psychosocial Health Needs. Washington, DC: National Academies Press; 2008.
- 6.↑
Commission on Cancer. Optimal Resources for Cancer Care: 2020 Standards. Accessed December 1, 2021. Available at: https://www.facs.org/-/media/files/quality-programs/cancer/coc/optimal_resources_for_cancer_ care_2020_standards.ashx
- 7.↑
Mehnert A, Hartung TJ, Friedrich M, et al. One in two cancer patients is significantly distressed: prevalence and indicators of distress. Psychooncology 2018;27:75–82.
- 8.↑
Herschbach P, Britzelmeir I, Dinkel A, et al. Distress in cancer patients: who are the main groups at risk? Psychooncology 2020;29:703–710.
- 9.↑
Apenteng BA, Hansen AR, Opoku ST, et al. Racial disparities in emotional distress among cancer survivors: insights from the Health Information National Trends Survey (HINTS). J Cancer Educ 2017;32:556–565.
- 10.↑
Massie MJ. Prevalence of depression in patients with cancer. J Natl Cancer Inst Monogr 2004;2004:57–71.
- 11.↑
John DA, Kawachi I, Lathan CS, et al. Disparities in perceived unmet need for supportive services among patients with lung cancer in the Cancer Care Outcomes Research and Surveillance Consortium. Cancer 2014;120:3178–3191.
- 12.↑
Ellis KR, Black KZ, Baker S, et al. Racial differences in the influence of health care system factors on informal support for cancer care among black and white breast and lung cancer survivors. Fam Community Health 2020;43:200–212.
- 13.↑
Yee MK, Sereika SM, Bender CM, et al. Symptom incidence, distress, cancer-related distress, and adherence to chemotherapy among African American women with breast cancer. Cancer 2017;123:2061–2069.
- 14.↑
Mausbach BT, Schwab RB, Irwin SA. Depression as a predictor of adherence to adjuvant endocrine therapy (AET) in women with breast cancer: a systematic review and meta-analysis. Breast Cancer Res Treat 2015;152:239–246.
- 15.↑
Kissane D. Beyond the psychotherapy and survival debate: the challenge of social disparity, depression and treatment adherence in psychosocial cancer care. Psychooncology 2009;18:1–5.
- 16.↑
Lazenby M, Ercolano E, Grant M, et al. Supporting commission on cancer-mandated psychosocial distress screening with implementation strategies. J Oncol Pract 2015;11:e413–420.
- 17.↑
Zebrack B, Kayser K, Padgett L, et al. Institutional capacity to provide psychosocial oncology support services: a report from the Association of Oncology Social Work. Cancer 2016;122:1937–1945.
- 18.↑
Kayser K, Brydon DM, Moon H, et al. Institutional capacity to provide psychosocial care in cancer programs: addressing barriers to delivering quality cancer care. Psychooncology 2020;29:1995–2002.
- 19.↑
Carlson LE, Waller A, Groff SL, et al. Online screening for distress, the 6th vital sign, in newly diagnosed oncology outpatients: randomised controlled trial of computerised vs personalised triage. Br J Cancer 2012;107:617–625.
- 20.↑
Vodermaier A, Linden W, Siu C. Screening for emotional distress in cancer patients: a systematic review of assessment instruments. J Natl Cancer Inst 2009;101:1464–1488.
- 21.↑
Pirl WF, Fann JR, Greer JA, et al. Recommendations for the implementation of distress screening programs in cancer centers: report from the American Psychosocial Oncology Society (APOS), Association of Oncology Social Work (AOSW), and Oncology Nursing Society (ONS) joint task force. Cancer 2014;120:2946–2954.
- 22.↑
Sutton TL, Koprowski MA, Grossblatt-Wait A, et al. Psychosocial distress is dynamic across the spectrum of cancer care and requires longitudinal screening for patient-centered care. Support Care Cancer 2022;30: 4255–4264.
- 23.↑
Barclay C, Viswanathan M, Ratner S, et al. Implementing evidence-based screening and counseling for unhealthy alcohol use with Epic-based electronic health record tools. Jt Comm J Qual Patient Saf 2019;45:566–574.
- 24.↑
Richardson LP, Zhou C, Gersh E, et al. Effect of electronic screening with personalized feedback on adolescent health risk behaviors in a primary care setting: a randomized clinical trial. JAMA Netw Open 2019;2:e193581.
- 25.↑
Satre DD, Anderson AN, Leibowitz AS, et al. Implementing electronic substance use disorder and depression and anxiety screening and behavioral interventions in primary care clinics serving people with HIV: protocol for the Promoting Access to Care Engagement (PACE) trial. Contemp Clin Trials 2019;84:105833.
- 26.↑
Pew Research Center. Internet/Broadband Fact Sheet. Accessed December 1, 2021. Available at: https://www.pewresearch.org/internet/fact-sheet/internet-broadband/#internet-use-over-time
- 27.↑
Nouri S, Khoong EC, Lyles CR, et al. Addressing equity in telemedicine for chronic disease management during the COVID-19 pandemic. Accessed December 1, 2021. Available at: https://catalyst.nejm.org/doi/full/10.1056/CAT.20.0123?targetBtn=articleToolSaveBtn
- 28.↑
Berkowitz SA, Traore CY, Singer DE, et al. Evaluating area-based socioeconomic status indicators for monitoring disparities within health care systems: results from a primary care network. Health Serv Res 2015;50:398–417.
- 29.↑
Krieger N, Chen JT, Waterman PD, et al. Choosing area based socioeconomic measures to monitor social inequalities in low birth weight and childhood lead poisoning: the Public Health Disparities Geocoding Project (US). J Epidemiol Community Health 2003;57:186–199.
- 30.↑
Sutton TL, Walker BS, Nabavizadeh N, et al. ASO visual abstract: geographic disparities in referral rates and oncologic outcomes of intrahepatic cholangiocarcinoma: a population-based study. Ann Surg Oncol 2021;28:716–717.
- 31.↑
Kwak SK, Kim JH. Statistical data preparation: management of missing values and outliers. Korean J Anesthesiol 2017;70:407–411.
- 33.↑
Herschbach P, Keller M, Knight L, et al. Psychological problems of cancer patients: a cancer distress screening with a cancer-specific questionnaire. Br J Cancer 2004;91:504–511.
- 34.↑
Walker ZJ, Xue S, Jones MP, et al. Depression, anxiety, and other mental disorders in patients with cancer in low- and lower-middle-income countries: a systematic review and meta-analysis. JCO Glob Oncol 2021;7:1233–1250.
- 35.↑
Berry DL, Blonquist TM, Patel RA, et al. Exposure to a patient-centered, web-based intervention for managing cancer symptom and quality of life issues: impact on symptom distress. J Med Internet Res 2015;17:e136.
- 36.↑
van den Berg SW, Peters EJ, Kraaijeveld JF, et al. Usage of a generic web-based self-management intervention for breast cancer survivors: substudy analysis of the BREATH trial. J Med Internet Res 2013;15:e170.
- 37.↑
Choi NG, Dinitto DM. The digital divide among low-income homebound older adults: internet use patterns, eHealth literacy, and attitudes toward computer/Internet use. J Med Internet Res 2013;15:e93.
- 38.↑
Blumenthal D, Tavenner M. The “meaningful use” regulation for electronic health records. N Engl J Med 2010;363:501–504.
- 39.↑
Stein JN, Klein JW, Payne TH, et al. Communicating with vulnerable patient populations: a randomized intervention to teach inpatients to use the electronic patient portal. Appl Clin Inform 2018;9:875–883.
- 40.↑
Ramsey A, Lanzo E, Huston-Paterson H, et al. Increasing patient portal usage: preliminary outcomes from the MyChart Genius Project. J Adolesc Health 2018;62:29–35.
- 41.↑
Beckjord EB, Rechis R, Nutt S, et al. What do people affected by cancer think about electronic health information exchange? Results from the 2010 LIVESTRONG Electronic Health Information Exchange Survey and the 2008 Health Information National Trends Survey. J Oncol Pract 2011;7:237–241.
- 42.↑
Bluethmann SM, Coa KI, Alfano CM, et al. Electronic health information exchange opportunities for self-management of care: responses from older adults with and without cancer history in the United States. Curr Oncol Rep 2018;20:30.
- 43.↑
Gonzalez BD. Promise of mobile health technology to reduce disparities in patients with cancer and survivors. JCO Clin Cancer Inform 2018;2:1–9.
- 44.↑
Berry DL, Hong F, Halpenny B, et al. Electronic self-report assessment for cancer and self-care support: results of a multicenter randomized trial. J Clin Oncol 2014;32:199–205.
