Patient reported factors influencing the decision‐making process of men with localised prostate cancer when considering Active Surveillance—A systematic review and thematic synthesis

Outcomes for men with localised prostate cancer managed with Active Surveillance (AS) are similar to outcomes for men who have received Active Treatment. This review explore men's perceptions of the factors that influence their decision‐making process when considering AS.


| INTRODUCTION
Prostate cancer is the second most common cancer diagnosed in men with approximately 1.1 million new cases diagnosed worldwide in 2012. 1 The cumulative lifetime risk of a diagnosis of prostate cancer for men in more developed countries is 12.3%, and in less developed countries is 4.4%. 2 Differences in lifetime risk between countries represent differences in the age of the population, and use of prostate specific antigen (PSA)-based screening for prostate cancer. 3 There was a rise in diagnoses of prostate cancer in the United States and Europe from the mid-1980s due to PSA testing. 4 However in 2012 the US Preventive Services Task Force issued guidelines recommending against PSA-based screening for prostate cancer for men of all ages due to false positives, overdiagnosis of prostate cancer, and overtreatment. 5 This recommendation led to an increased awareness of overtreatment of prostate cancer and resulted in an increase in Active Surveillance (AS). 6 Analysis of cancer registry data in the US indicates that AS rates for localised prostate cancer have increased between 2010 and 2014 from 30.2% to 57.5%, however there are differences in rates of AS due to geography, age, marital status and insurance status. 7 Being older, unmarried and uninsured were all independent predictors of choosing AS. 7 The European Association of Urology (2021) guidelines state that over-treatment is the main risk for men with low-risk prostate cancer, and recommend that AS should be considered for all low-risk patients. 3 Both individual and physician factors have been found to influence the decision to choose AS for prostate cancer. 8 Kinsella et al. (2018) conducted a mixed methods review of patient factors influencing choice of AS and found that patient and tumour factors, family and social support, provider attitudes, type of healthcare organisation and health policy all influenced choice. 9 However, the Kinsella review only included qualitative studies with more than 20 participants, and conducted a content analysis of the qualitative studies, identifying barriers and facilitators to choosing AS. 9 Qualitative evidence synthesis allows for the analysis of complex qualitative data to gain a richer and more detailed understanding of men's experiences, attitudes and priorities for decision making about their prostate cancer treatment. This systematic review will therefore use thematic synthesis to explore men's perspectives on the factors that influence their decision-making process when considering AS.

| METHODS
This review was prospectively registered with PROSPERO (CRD42020197686) and reported according to the PRISMA checklist and ENTREQ statement (Table S1, Table S2). 10,11

| Research question
What factors influence the decision-making process of men with localised prostate cancer when considering AS?

| Search strategy
Comprehensive pre-planned searches of published qualitative literature were conducted in July 2020, using EBSCO CINAHL, Medline -Pubmed, Embase, Psychinfo and SCOPUS. Searches were updated in May 2021. Papers to be included in the review were identified using the sample, phenomenon of interest, design, evaluation, research type (SPIDER) tool. 12 Papers were included if they reported qualitative data about men's decision-making process around considering (choosing or not choosing) AS for treatment of localised prostate cancer. Papers were excluded if they reported single case studies, quantitative and mixed methods studies unless the qualitative data could be separated from the quantitative data, studies not published in English, full text not available, or studies which focused on partners', couples', or health care professionals' decision-making Entire results sections, comprising quotations and interpretation of included papers were extracted into QSR NVIVO version 12.

| Quality appraisal
Two review authors (Maggie Cunningham and Mike Murphy) independently assessed the methodological and reporting quality of included studies using the critical appraisal skills programme qualitative checklist. 13 Consensus was reached on the quality of included studies through discussion. No studies were excluded on the basis of the quality appraisal.

| Data synthesis
Data was analysed using thematic synthesis. 14 In the first stage of the synthesis, all first and second order data from the Results sections of included studies was coded line by line according to its meaning and content, to search for concepts. Coding was completed in QSR NVIVO version 12. The papers were coded in alphabetical order, and the codes from the first paper were translated into the second paper and so on until all papers had been coded. Coding was therefore an evolving, inductive and iterative process, and as new codes were generated previous papers were re-visited to ensure that all instances of that

| Study selection
A total of 79 unique papers were identified. After screening of abstracts, 31 papers remained and the full texts of these papers were assessed for eligibility against the inclusion criteria. Eighteen further papers were excluded after full-text screen; reasons for exclusion were that the studies were not about decision making, no qualitative results were provided, relevant data could not be separated from other data, or the same data was included in multiple papers. The PRISMA flow diagram for the number of papers included and excluded at each stage is shown in Figure 1. 10 Study characteristics of included studies are presented in Table 2.

| Study quality
There was some variation in the comprehensiveness of reporting of included studies. Only one study provided information on all of the CASP checklist items. 16 The CASP checklist item which was most poorly reported was 'has the relationship between researcher and participants been adequately considered?' In general, no or little information was provided about the relationship between the researchers and participants or the impact of the researchers on question formulation, conduct of the interviews or the analysis. CASP item 'was the data analysis sufficiently rigorous?' was also poorly reported in five papers. In those papers analysis was only briefly described, and it was not clear how themes were derived from the 390data. Details of the CASP quality appraisal for each included study are provided in Table 3. No studies were excluded on the basis of the quality assessment. denotes an author comment.

| Pre-diagnosis representations of cancer and treatment
The Attitudes and beliefs about AS; 3. Reasons for rejecting AS; -relative 'youth'; -'doing nothing' is not an option;anxiety and fear of cancer spreading; -persuasion against AS Abbreviation: AS, Active Surveillance.

-T A B L E 3 CASP quality appraisal summary table
Bellardita et al. 15 Berger et al. 16 Chapple et al. 17 Davison et al. 18 Fitch et al. 19 Holmboe et al. 20 Lyons et al. 21 Mallapareddi et al. 22 O'Callaghan et al. 23 Parker et al. 24 Seaman et al. 25 Volk et al. 26 Xu et al. 27 Was    But if I understand right it's not that surgery is going to completely eliminate the problem in the long run.
That's what I was told, it could come back.
[P] 15 What persuaded me most was the reaction of the medical staff. They didn't seem to be overly excited about the whole thing.

Doctor and system factors:
[P] 19 I feel like because I was a black man that they were willing to use me as a guinea pig and if they could make some money… [P] 22 Men who chose AS relied on robust protocols and procedures within the healthcare system. Trusting in these protocols, for example, having regular appointments for biopsies, blood tests, scans and consultations, allowed them to off-set the risk of continuing to live with cancer inside them. Men were reassured by the protocols for regular monitoring, and needed to believe that the regular checks would quickly pick up any progression in their cancer. Some men lost faith in the reliability of PSA or biopsy results, due to inconsistencies in results or because of information they had read in research or on the internet. This was a source of anxiety for men and a reason why some opted for Active Trust in the doctor, and the doctor's opinions and advice regarding treatment options were very influential in whether or not men chose AS. Men who were not offered AS as an option by their doctor rarely pursued it as an option. Heuristic processing is thought to reduce effort and increase efficiency in decision making through using commonsense rules of thumb to make decisions based on a subset of available information. 28 Heuristic processes include expert opinion heuristics, where decisions are guided by the rule that 'experts can be trusted'. 28 Steginga (2004) found that men with a low tolerance of ambiguity were more likely to use expert opinion heuristics to aid decision making, and that when expert opinion heuristic was used, systematic information processing decreased. 28 The findings of this review suggest that men find synthesising information about prostate cancer challenging and therefore unhelpful for decision making. This is in line with previous research which has found that men do not use information comprehensively or systematically when making treatment decisions, and instead rely on their previous lay beliefs about cancer and health. 29 These findings highlight the importance of the clinician and tailored information to decision making.
Previous research into decision making of men with localised prostate cancer found that AS enables men to put off making a decision about Active Treatment for the time-being, and this was seen as a major benefit of choosing AS. It allowed men to avoid the negative side effects of Active Treatment for as long as possible. Anticipated regret has been found to influence decisionmaking processes-higher levels of anticipated regret from engaging in a behaviour predict weaker intentions and behaviour, whereas greater anticipated regret from not engaging in a behaviour predicts stronger intentions and behaviour. 30 There is the potential for anticipated regret in each treatment choice-in choosing AS, the anticipated inaction regret may be that the cancer will progress leading to worse outcomes in the long term; in choosing Active Treatment, the anticipated action regret may be the negative side effects of treatment. Of these potential antici-

| Study limitations
The papers included in this study were heterogeneous, asking different research questions, using different methods of analysis, and including participants with varied inclusion criteria. In translating findings from one study to another, the specific context of the individual studies can be compromised -a problem inherent to qualitative synthesis. The studies included in this review were mainly conducted in Anglophone countries (USA, UK, Canada, Australia), reducing the generalisability of findings. However, the themes identified in this thematic synthesis were also supported by the study conducted in Italy, providing tentative support for generalisability of the findings to other countries which have AS treatment protocols.
Our review identified that the studies included did not tend to distinguish between AS and watchful waiting. We also note that the studies spanned a time period from 2000 to the present day, which may reflect changing attitudes to and protocols for AS.