Patient-reported outcomes and psycho-oncological screening in hematology: a practical example of routine electronic monitoring

PRO assessments are typically used in clinical studies or in routine care monitoring. On a basic level, PROs on QoL are either broad instruments applicable to a general population, or specific instruments validated for certain groups (e.g., disease groups like cancer). Usage of generic PRO tools like the EQ-5D or the Short Form Health Survey (SF-36) permits a comparison across different disease groups. However, those questionnaires do not cover disease-specific domains and symptoms that may be relevant to patients. In most cases, using a cancer-specific measurement is advisable, as those more precisely measure issues relevant to oncological patients. One of the most widely used PRO measures is the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire (EORTC QLQ-C30) [8‐10], developed by the EORTC Quality of Life Group, which can be supplemented using disease-specific modules for different types of cancer. Other cancer-specific PRO tools are the PROMIS-Cancer measures by the National Cancer Institute (NCI) or the Functional Assessment of Cancer Therapy – General (FACT‑G). Depending on the focus of the assessments, the PRO measure should be selected carefully [11, 12]. Additional specific questionnaires can be used to assess particular domains of interest such as the need or indication for psycho-oncological treatment, e.g., Distress Thermometer [13] or Hornheider Screening Instrument (HSI) [14].

While in clinical studies the results of PRO measures are often analyzed after conducting the study, requirements for PRO usage in clinical routine are different. Research has shown that the integration of PRO results into the clinical workflow benefits from instant scoring and comprehensive PRO result presentation [15, 16]. This can be accomplished with electronic patient-reported outcome (ePRO) data collection. Using PCs, tablet-PCs or smartphone devices, patients can report on PRO measures. Directly after completion, HCPs can review the results and tailor the clinical encounter to the patient’s responses. Compared to paper–pencil data collection, ePRO data collection has been shown to improve data integrity and reduce missing data in clinical trials [17] while offering the same validity [18].

Assessing PROs offers a variety of benefits in the clinical setting. PROs can systematically assess relevant symptoms and side effects that may be relevant to treatment. The PRO results, especially when collected electronically, can help HCPs gain a quick and comprehensive impression of the patients’ perceived health. Review of the results may lead to clinical exploration of the data or initiate clinical action. For example, for the EORTC QLQ-C30, thresholds for clinical importance are available indicating a high possibility for clinically important problems in a patient’s QoL [19]. These thresholds can be used for symptom screening in clinical practice. Well-validated and established PRO tools like the EORT QLQ-C30, combined with interpretation guidelines such as normative data or thresholds for clinical importance, allow the comprehensive and systematic assessment and interpretation of patients’ QoL. Due to the broad assessment focus of standardized questionnaires, different sensitive domains (like fatigue, sexuality) are addressed which are otherwise less likely to be picked up during consultation [20‐22]. At the same time, a wider variety of topics can be covered in the consultations if PRO results are used because PROs can cover a broad range of QoL domains [23, 24]. Another welcome effect is that PRO results can support focused dialogue between HCPs and patients, as they give an indication on which topics or QoL domains merit in-depth discussion [20]. Even though more topics are picked up and discussed, the time spent with the patient during consultation remains the same [20, 24]. Focusing patient-relevant issues and evaluating patient priorities can support shared-decision making in therapy which in turn can increase treatment satisfaction and adherence in patients [24‐26].

In the last few years, the first RCT studies were able to show concrete survival benefits of PRO usage in the oncological setting. Basch et al. [6] report on advanced lung cancer patients using a weekly web-based symptom monitoring during routine care. In the usual care group, no ePRO monitoring was carried out. If a patient in the ePRO group deteriorated or showed specific symptoms, an HCP was notified and could update the treatment accordingly (e.g., schedule additional imaging). Patients in the ePRO monitoring group showed a median prolonged survival of 5 months compared to patients in standard care. Similar results were reported by Denis et al., recruiting advanced nonprogressive stage IIA (TXN1) to IV lung cancer patients at five French centers [7]. In their study, median survival was 22.5 months (ePRO group) versus 13.5 months (standard of care). The authors conclude that ePRO integration enabled more adequate and timelier provision of care and therefore prolonged survival. Early detection of and adequate response to side effects are known factors to reduce unplanned hospital visits and increase patients’ therapy adherence [27]. Furthermore, patients in the study by Denis et al. also tolerated longer chemotherapy than patients in usual care, which could be another contributing factor.

A cancer diagnosis is often physically and emotionally burdensome for patients and can result in high psychosocial and emotional distress for the individual and/or family [34]. According to the literature, 20–50% of cancer patients are emotionally distressed [35, 36] or have psychiatric comorbidities during therapy [37]. Consequently, cancer treatment should be supplemented with adequate psychosocial support such as psycho-oncological treatment which can increase patients’ QoL and decrease distress [38, 39]. For HCPs, it can be challenging to identify patients in need of support and referral to psycho-oncological services as many patients do not proactively report distress or seek help, and HCPs might lack specific training to identify distressed patients [40, 41]. Although cancer-specific questionnaires usually assess emotional functioning or distress, this does not suffice for the identification of patients in need of psycho-oncological support and treatment [42]. Therefore, treatment guidelines suggest the use of a routine psycho-oncological screening tool [43]. In August 2019, the Hornheider Screening Instrument (HSI) [14] was added to the ePRO routine monitoring at our unit. The original version of the HSI contains seven items assessing emotional strain, unmet information needs and disease burden in the family. A cut-off indicates patients who would likely benefit from psycho-oncological treatment. For the screening of psycho-oncological treatment needs at the Department of Hematology Innsbruck, the HSI is supplemented with five additional items on episodes of depressive mood or anxiety, previous psychological or psychiatric treatment, and patients’ personal wish to talk to a psycho-oncologist. This results in a total of 12 items which has been shown to be feasible for screening purposes [44]. To keep response burden low, patients complete the HSI screening in 3‑month intervals. In CHES, the HSI cut-off is linked to a graphical indicator (a blue exclamation mark) in the patient overview, which makes it easy for HCPs to identify patients who might benefit from psycho-oncological treatment. The exclamation mark is displayed when patients exceed the HSI cut-off (regardless of whether they wish to see a psycho-oncologist) or when they express the wish to see a psycho-oncologist (regardless of whether they exceed the cut-off). Fig. 2 (adapted from [45]) displays the possible outcomes and resulting actions taken by the staff depending on the HSI results.

Between implementation of the screening in August 2019 and January 2020, 91 patients completed at least one HSI screening. Of those, 40 (44%) exceeded the cut-off during the first screening but only 5 (5%) wished to talk to a psycho-oncologist. This gap between experienced distress and expressed wish for psycho-oncological support is known from other studies, although normally less prominent (e.g., [46‐48]). Proposed reasons for few patients expressing the wish to see a psycho-oncologist are the fear of social stigma, a lack of knowledge about psycho-oncology, a preference for self-help, and the belief that their own distress is not severe enough to merit intervention [46, 49].

The patient perspective on the PRO assessments is a critical factor for any implementation. Research suggests that how patients experience the assessments is connected to their compliance [50] and satisfaction with the PRO usage [51]. At our outpatient unit, patients with MM and CLL are approached in the clinic by the ePRO facilitator. They are asked to participate in the monitoring program and, in case of a MM diagnosis, to consent to data usage by the AMR. After providing informed consent, the patient is presented further information about the goals and procedure of the monitoring. Patients then fill in the questionnaires for the first time on a tablet computer in a designated room. Patients can choose their preferred or native language, which reduces misunderstandings and includes foreign-language speakers. Due to the multilingual assessments, ePROs can help bridge language-based communication gaps between patients and physicians as they provide an extensive assessment in various languages. During questionnaire completion, the ePRO facilitator is present and can assist patients who require support with the assessment. The assessment is usually carried out between the blood sampling and the appointment with the doctor. This ensures that ePRO results can be reviewed by the doctor before and during the consultation. After the first inclusion into the routine monitoring, regular assessments are carried out at each subsequent visit at the outpatient unit. Assessment intervals differ with appointment frequency and can range anywhere between 2 weeks and 12 months.

Patients who are eligible to participate in an electronic PRO home monitoring (at least some internet usage, and no overt cognitive impairment) are introduced to our patient portal. Approval rates for the portal are high: most eligible patients agree to use the portal (79 of 88; 90%). Aside from questionnaire completion, the patient portal offers additional features for the patients. It provides reliable disease-specific information composed by the physicians and content which is dependent on patients’ questionnaire results. Self-management tools are dynamically displayed for QoL domains if a threshold for clinical importance is reached [19]. The self-management tools consist of advice and instructions on how to handle disease- and treatment-related symptom burden (e.g. breathing exercise, dietary information), or contact information to relevant counseling sources such as psycho-oncology. Patients are recommended to view the self-help tools for domains in which they reached the predefined threshold. Moreover, patients can view their own results and compare them to reference data from other patients, a wish frequently reported by cancer patients [52].

Physicians entering the program for the first time are also introduced to the routine monitoring by the ePRO facilitator and are provided their own login data for the CHES system. Their account allows them to log into the CHES platform using just an internet browser at any time. Evaluation of the feasibility in our unit shows that the effort and burden of the ePRO assessments for physicians is low [28]. Typically, physicians are advised to review their patient’s PRO results right before or during the consultation. They can choose just how much specific information they want to extract from the ePRO assessments: (level 1) inspect just a pie chart summarizing the results of all subscales of the three questionnaires and an indicator of change from the last assessment, (level 2) inspect the results for all subscales split by functions and symptoms, or (level 3) to check the specific answers a patient gave that led to the results at hand. This way physicians can adapt the amount of specific information to their needs. For instance, if the patients reported QoL did not deteriorate, it might be sufficient to just check the summarizing pie chart (Fig. 3). Otherwise, if the patient shows deterioration in some domains, the doctor can inspect the respective items and receive detailed information.

Routinely assessed PRO data can support HCPs in more easily adapting a patient-centered approach. Thresholds for clinical importance are used to identify QoL domains that might be clinically relevant [19]. If the threshold is passed for a subscale, it is likely that the patient experiences clinically important problems in this domain. Physicians are advised to discuss those domains with the patient. The blue exclamation mark as an indication of a possible psycho-oncological treatment indication helps in the early recognition of patients who might benefit from further counselling and triggers adequate clinical action. Moreover, as the psycho-oncologists have access to the QoL data, they can obtain additional information on the health state of the patient from the PRO results before even meeting him or her.