| Abstract | Treatment adherence in cancer is crucial to health outcomes - however medication nonadherence is often in the region of 50% and has shown little change in over two decades. Although studies consistently report the extent of the problem, there are limited qualitative explanations. Additionally, the increasing number of oral medications and subcutaneous therapies self-administered away from the clinical setting may have exacerbated the risk of medication nonadherence. Studies that assess or report treatment nonadherence are often from a monolithic concept standpoint and a medication compliance perspective. This narrow view exists despite increased conceptual clarity provided by the World Health Organisation (WHO) that define the phenomenon as inter-dimensional and multi-factorial (Sabaté, 2003, p. 114). Although the WHO definition of treatment adherence is centred around a patient dimension - physicians are not similarly highlighted but absorbed into a multi-factorial health system / health care team dimension. This lack of distinction may have placed greater emphasis on the cancer patient for treatment adherence and less focus on physician communication and information. In the context of oncologist relationships during the cancer patient’s journey, studies have shown that the communication factor may be as significant as a medical intervention in health outcomes. This thesis focuses upon the unmet informational needs of cancer patients and the potential influence of oncologist communication in treatment nonadherence.
In method triangulation, an initial systematic review has first explored and defined nonadherence concepts. This is followed by a scoping review of natural language processing techniques to evaluate the cancer patient’s lived experience. In a third phase, qualitative validation interviews are undertaken with cancer patients and oncologists. The actionable evidence of modifiable factors in oncologist communication and information needed by patients is derived from qualitative findings in several key areas of this research; a) oncologist to patient communication factors that may influence intentional or unintentional treatment nonadherence; b) The cancer patients’ lived experience and inferences or conceptual associations that may suggest or infer treatment nonadherence; c) The oncologist perspective of communication and how medical narratives (notes) in Electronic Health Records could enhance treatment concordance; d) Cancer patient online support communities and their impact upon unmet informational and emotional needs of patients.
Features considered essential in framework development of an online intervention to support the cancer patient’s unmet communication and informational needs are included in this research. Underpinning this research are key aims of patient and person centred care (PCC) where the preferences of patients are central to care and there is shared decision making and agreement (concordance) on treatment recommendations. PCC is somewhat aligned with the WHO holistic definition of treatment adherence and a transition from an outdated paternalistic approach to one that is both patient and person centred. However, evidence from method triangulation in this research, suggests that despite a paradigm shift to PCC, oncologists may practice a medication compliance approach.
Findings have shown that traditional studies into patient satisfaction typically use a survey design. These are often developed from a clinical needs’ perspective, with an emphasis on Health-Related Quality of Life (HRQOL). This does not entirely represent Quality of Life (QOL) and the daily living priorities that cancer patients may have. This knowledge gap has led to the need for a deeper understanding of the cancer patients’ reality or lived experience. Therefore, in method triangulation a scoping review was performed of diverse studies that use natural language processing (NLP) techniques (primarily, sentiment analysis) to assess the cancer patient’s lived experience. A hypothesis of this research is that NLP studies into the cancer patient’s lived experience can also provide rich adherence related data. This information may provide knowledge of treatment nonadherence that is hidden from traditional patient self-reports and other quantitative surveys. Due to a paucity of NLP studies that report on cancer patient treatment nonadherence, the above-mentioned strategy was performed in a scoping review with a narrative and discourse analysis approach. The scoping review explores the cancer patient’s lived experience for possible inferences or associations with nonadherence concepts and themes identified in the initial systematic review. To gain insight into the oncologist perspective of adherence, the scoping review also explores the use of text mining - an NLP technique used to analyse medical notes or narratives in Electronic Health Records (EHRs). Studies that use text mining of oncologist notes in the specific context of treatment adherence or nonadherence are limited; Due to the scarcity of adherence related data in oncologist narratives, information has been supplemented by findings from several qualitative surveys into oncologist communication and barriers to use of EHRs.
In an initial systematic review, the principles of meta-ethnography have been applied and a line of argument formed. The overarching aim of the systematic review is to explore when treatment nonadherence may be intentional or unintentional. A hypothesis of this research is that knowing this differentiation can improve knowledge of oncologist - cancer patient relationships, and future framework development of interventions to enhance treatment adherence. Among broad findings from the systematic review are that unintentional nonadherence frequently overlaps with intentional nonadherence, and that the influence of oncologist communication is often missing or understated.
Evidence from this research suggests that despite the transition to patient centred care (PCC) a medication compliance approach is often practiced. In a similar context studies appear to focus on medication compliance and rarely treatment adherence as a multi-factorial concept. Qualitative validation interviews provide evidence that patients frequently seek clarification and further understanding of treatment recommendations from participation in online community peer support groups and other online resources. Evidence from method triangulation has shown that cancer patient participation in online support communities can provide both emotional and informational support - however, somatization may also be a risk factor.
Findings from this thesis often relate to modifiable oncologist factors in relation to communication and information. Qualitative validation interviews suggest that oncologists may view medication nonadherence differently, e.g. missing a dose was not necessarily viewed as nonadherence. Patient access to Electronic Health Records (EHRs) offers the potential for cancer patients to read the notes (medical narratives) made by oncologists in EHRs. This may stimulate engagement with oncologists and lessen the risk of misinformation from cancer patient participation in online support groups. Evidence from method triangulation in this Thesis suggests that patient access to EHRs, and informed communication that follows has the potential to improve treatment concordance and enhance treatment adherence. |
|---|