Challenges in the cultural adaptation of the German Myeloma Patient Outcome Scale (MyPOS): an outcome measure to support routine symptom assessment in myeloma care

With the ageing of society and an increasing incidence [1‐3], cancer is a major public health concern. Haematological cancers and multiple myeloma (MM) in particular exemplify this changing face of cancer with conditions whose management resembles that of a chronic illness, with recurrent treatment patterns followed by maintenance therapy [4, 5]. MM is an incurable cancer of the bone marrow characterised by bone destruction, bone marrow failure with anaemia, immune deficiency, and renal insufficiency [6]. It belongs to the heterogeneous group of plasma cell dyscrasias, which vary from asymptomatic forms to malignant disease with severe end-organ damage and high patient morbidity [7]. Front-line treatment with high-dose chemotherapy and hematopoietic stem cell transplant (HSCT) has improved the median survival for those under the age of 65 to 5 years or longer after a myeloma diagnosis [8]. However, due to its incurability, patients must cope with the incrementally progressive nature of the disease, interspersed with intervals of stable disease and maintenance treatment, while also experiencing long-lasting effects of previous treatments [9, 10]. Moreover, the median age at diagnosis is 69 years, with more than 60% of multiple myeloma patients aged 65 years or older at diagnosis. Due to the aging baby boomer generation, an increase in the proportion of the general population older than 65 years of age is expected between 2020 and 2030. Therefore, in tandem with new and better treatment options, the incidence and prevalence, and subsequently also deaths due to multiple myeloma, will inevitably increase [3].

Therefore, MM patients represent a group of older, haematological cancer patients faced with complex treatment pathways, a long duration of a chronic, yet life-threatening disease with repeated relapses and high levels of uncertainty around disease and treatment progression [11‐13]. This results in a high prevalence of physical and psychosocial symptoms and problems throughout the disease trajectory. MM patients may suffer more symptoms and problems than other patients with haematological malignancies. On average, several cross-sectional surveys have shown a mean of 5.6 symptoms, of which 2.3 were rated as severe [9]. A recent meta-analysis showed high prevalence rates for pain and fatigue as well as limitations in aspects of quality of life (QoL) such as physical, role, and social functions [14].

Haematological and psychosocial care of MM patients could be improved by incorporating a longitudinal assessment of symptoms and QoL into routine clinical practice. The routine use of QoL measures allows monitoring of symptoms/QoL, thus leading to better symptom control, improved communication, and higher patient satisfaction [15, 16]. However, few measures have been designed for monitoring QoL in the routine clinical setting; a systematic review of 13 generic and disease-specific health-related QoL measures for multiple myeloma found no single tool developed or validated specifically for routine clinical care [17]. The most common measures, the Functional Assessment of Cancer Therapy-Multiple Myeloma questionnaire (FACT-MM) [18], the EORTC QLQ-MY20 [19], and the M. D. Anderson Cancer Centre Multiple Myeloma measure (MDASI-MM) [20] are disease-specific measures that have been developed for use in clinical trials. Currently, only the EORTC QLQ-MY20 is available in German [21].

The Myeloma Patient Outcome Scale (MyPOS) is a questionnaire developed and validated specifically to measure disease-specific QoL in patients with MM in a routine clinical setting and for monitoring purposes. The initial validation comprised 380 patients at different disease stages in the UK [22]. The MyPOS is a module of the Integrated Palliative/Patient care Outcome Scale (IPOS) [23‐25], a short, multidimensional questionnaire to assess palliative care concerns in patients with advanced disease. The MyPOS takes the core items of the POS and extends them with myeloma-specific concerns. The MyPOS comprises a list of 13 symptoms and 20 QoL items which are scored on a 5-point Likert scale and summed into a total score and three subscale scores. Content and construct validity as well as reliability of the MyPOS have been established in clinically representative samples of all disease stages [22, 26, 27]. Its reliability and responsiveness in longitudinal monitoring of QoL have been shown to be satisfactory to good [27].

However, the conceptualization of QoL for patients with MM used in the original MyPOS is UK-focused and must be checked before a German translation of the instrument can be used in practice. Differences in care processes, treatment pathways and cultural background may affect its cross-cultural equivalence [28, 29]. We therefore aimed to explore the cultural, linguistic, and contextual issues during the adaptation of the MyPOS to the German context. Our aim was to translate and cross-culturally adapt the MyPOS and establish its content and face validity for monitoring QoL in MM patients.

This study used a multi-step, explorative and sequential mixed-methods design [30, 31]. The procedure for the translation and cultural adaptation of the MyPOS followed the guidelines reported in the manual for cross-cultural adaptation and validation of the parent’s measure POS development group [32], based on commonly accepted standards for cross-cultural adaptation and psychometric testing by the European Organization for Research and Treatment of Cancer (EORTC) and the International Society for Pharmacoeconomics and Outcomes Research (ISPOR) [33, 34]. Because there are already German translation versions of the POS and the IPOS and therefore challenges with adapting items to the German language and healthcare context are already known, we tailored the proposed, six-step process towards supporting a wider exploration of issues of cultural equivalence instead of focusing on psychometric evaluation (see Fig. 1). The cultural adaptation of MyPOS was further informed by the Cultural Equivalency Model for Translating and Adapting Instruments [35], focusing on conceptual (Does the content relate to constructs in the culture?), content (Is the content of each item relevant?), semantic (Is the meaning of the item’s wording the same?), and technical equivalence (Do layout, format, and answer options work the same way?).

In this study, we translated and culturally adapted the Myeloma Patient Outcome Scale, a patient-reported symptom and QoL tool to support monitoring of patient-centred issues in routine clinical practice. The MyPOS is the first disease-specific myeloma questionnaire available in German that supports both clinical assessment/monitoring as well as use within research studies. The only other available translation of a health-related QoL questionnaire, the EORTC QLQ-MY20 only offers utility within research [21]. The German translation of the MyPOS has been shown to possess content/face validity and acceptability for patients and staff. However, certain adaptations honouring patient-centred wishes for more or less information, separating patient satisfaction from QoL, and involving family members in the assessment process were needed to achieve cultural equivalence within a German healthcare context. The questionnaire was perceived as clinically important in preparing and guiding individual clinical encounters with HCPs and helping patients raise embarrassing or challenging issues with their HCPs. MyPOS may help to direct and focus conversations when employed longitudinally. The sequential use of focus groups to establish conceptual equivalence and cognitive interviews to cognitively test individual items benefitted the content validity by stressing separate issues regarding the comprehension and utility of the questionnaire.

Several of the issues regarding comprehension and equivalence of individual items observed in this study have also been reported in recent cultural adaptation studies of the parent measure, the IPOS. Our study also found more problems with the adaptation of items from the parent measure than from the myeloma-part of the measure. Comprehension issues consistently reported in the literature concern item Q5, feeling depressed. The association of the term ‘depressed’ with a medical diagnosis was also noted in a study testing a Swedish version of the IPOS [48]. Similar to our finding, the Swedish research group opted for a colloquial term. They found that this translation was better fitted towards eliciting the patient’s mood. Some authors argue for directly asking patients whether they think they are depressed rather than opting for a wording of feeling depressed. However, results are inconsistent [49, 50]. Similar findings have been reported in cognitive interviews during the German translation of the IPOS [25] and its Italian translation [51]. The issue around culturally adapting Q9 as well as Q19, both referring to problems of a financial or personal nature being addressed, was also reported in the French and Italian translation of Q9 in the IPOS [51, 52]. Additional problems concerning the response options have been described.

Items in the healthcare support subscale caused the most concerns from a content and face validity point of view. Patients identified issues regarding these questions measuring a construct other than QoL and issues regarding wording, adaptation and the global nature of these questions. Both the cross-sectional and longitudinal validity studies of the English MyPOS [22, 26] reported poor psychometric criteria for these items, with pronounced ceiling effects in the scale and poor reliability. Similar issues were found for a set of healthcare satisfaction items that were part of the original 24-item version of the EORTC QLQ-MY20 [53]. Due to the same reasons, these items were removed from subsequent versions of the questionnaire. Despite patient satisfaction being assigned a central role in patient outcomes [54], it is regarded as a patient experience measure rather than an outcome measure and thus perceived as distinct from QoL as a construct [55]. However, healthcare satisfaction and QoL do seem to overlap. The addition of these items to the original MyPOS was based on findings from the qualitative interviews to develop the measure [38]. In stem cell transplant populations, it has also been reported that patients with higher levels of satisfaction with medical care reported higher levels of QoL, despite ongoing physical and psychosocial morbidity after transplant [56]. The main critique in this German sample centred on worries of negative satisfaction statements sending the wrong message to HCPs and jeopardizing the long-lasting relationships with doctors and nurses who care for these patients throughout their illness trajectory. Therefore, the negative connotation of these items might well point towards different care models and patient experiences with care in the German context.

This less fragmented care model that was reported by patients in our study also serves the clinical utility of the MyPOS. Unlike the MDASI-MM and the EORTC-QLQ-MY20, the MyPOS contains more items regarding worry about the future, information needs, and coping processes as well as adaptation processes, all relevant to and reflecting the prolonged disease trajectory. These issues have also been highlighted as important in recent qualitative studies focusing on the advanced myeloma population [11‐13, 57, 58]. The MyPOS was developed as a tool to go beyond the simple assessment of symptom status and asking for an evaluation of physical, emotional, spiritual, and QoL morbidity. Outcome measures to be used in routine clinical care have been shown to be powerful instruments to improve wellbeing and outcomes of cancer patients as well as improve communication with HCPs [15]. These tools cannot replace the interventions required to meet patients’ and their family caregiver’s needs [59, 60], but they can help to identify and monitor high levels of unmet supportive care needs and help integrate those services into care [61, 62]. The interface of supportive care and PC with haematology differs from the one with oncology because MM patients usually have a long and close relationship with the haematology team. LeBlanc & El-Jawahri even proposed unifying PC with haematology [63] by having haemato-oncological staff with a qualification in PC follow a primary PC approach supported by close collaboration with PC specialists when required, as indicated e.g. by patient-reported outcomes and joint case reviews. Porta-Sales et al. demonstrated the benefit of a cooperative PC-haematology approach for outpatients guided by impaired QoL in a retrospective analysis of their Multiple Myeloma Palliative Care Clinic [64]. The MyPOS could support such a care model.

In this study, we translated and culturally adapted the Myeloma Patient Outcome Scale, a QoL questionnaire to support the assessment and monitoring of symptoms and wellbeing in routine clinical care. After adaptations to the German care context, the MyPOS demonstrated good face and content validity. MyPOS German is now ready to support successful clinical approaches to better meet myeloma patients’ needs for palliation of symptoms as well as psychosocial concerns which impact on QoL. Next steps in its German validation include formal psychometric and clinical utility testing. Further, the demand for a proxy version should be appreciated and addressed, including assessment of informal caregiver burden.