Article Text

Improving the quality of self-management support in ambulatory cancer care: a mixed-method study of organisational and clinician readiness, barriers and enablers for tailoring of implementation strategies to multisites
  1. Doris Howell1,
  2. Melanie Powis2,
  3. Ryan Kirkby2,
  4. Heidi Amernic3,
  5. Lesley Moody2,
  6. Denise Bryant-Lukosius4,
  7. Mary Ann O'Brien5,
  8. Sara Rask6,
  9. Monika Krzyzanowska7
  1. 1 Supportive Care, Princess Margaret Cancer Centre Research Institute, Toronto, Ontario, Canada
  2. 2 Medical Oncology and Hematology, Princess Margaret Hospital Cancer Centre, Toronto, Ontario, Canada
  3. 3 Cancer Care Ontario Branch, Ontario Health, Toronto, Ontario, Canada
  4. 4 Oncology, Hamilton Health Sciences Centre, Hamilton, Ontario, Canada
  5. 5 Family Medicine, University of Toronto, Toronto, Ontario, Canada
  6. 6 Medical Oncology, Royal Victoria Regional Health Centre, Barrie, Ontario, Canada
  7. 7 Medical Oncology & Hematology, Princess Margaret Hospital Cancer Centre, Toronto, Ontario, Canada
  1. Correspondence to Dr Doris Howell, Supportive Care, Princess Margaret Hospital Cancer Centre, Toronto, Canada; Doris.Howell{at}uhn.ca

Abstract

Introduction Improving the quality of self-management support (SMS) for treatment-related toxicities is a priority in cancer care. Successful implementation of SMS programmes depends on tailoring implementation strategies to organisational readiness factors and barriers/enablers, however, a systematic process for this is lacking. In this formative phase of our implementation-effectiveness trial, Self-Management and Activation to Reduce Treatment-Related Toxicities, we evaluated readiness based on constructs in the Consolidated Framework for Implementation Research (CFIR) and Normalisation Process Theory (NPT) and developed a process for mapping implementation strategies to local contexts.

Methods In this convergent mixed-method study, surveys and interviews were used to assess readiness and barriers/enablers for SMS among stakeholders in 3 disease site groups at 3 regional cancer centres (RCCs) in Ontario, Canada. Median survey responses were classified as a barrier, enabler or neutral based on a priori cut-off values. Barriers/enablers at each centre were mapped to CFIR and then inputted into the CFIR-Expert Recommendations for Implementing Change Strategy Matching Tool V.1.0 (CFIR-ERIC) to identify centre-specific implementation strategies. Qualitative data were separately analysed and themes mapped to CFIR constructs to provide a deeper understanding of barriers/enablers.

Results SMS in most of the RCCs was not systematically delivered, yet most stakeholders (n=78; respondent rate=50%) valued SMS. For centre 1, 7 barriers/12 enablers were identified, 14 barriers/9 enablers for centre 2 and 11 barriers/5 enablers for centre 3. Of the total 46 strategies identified, 30 (65%) were common across centres as core implementation strategies and 5 tailored implementation recommendations were identified for centres 1 and 3, and 4 for centre 2.

Conclusions The CFIR and CFIR-ERIC were valuable tools for tailoring SMS implementation to readiness and barriers/enablers, whereas NPT helped to clarify the clinical work of implementation. Our approach to tailoring of implementation strategies may have relevance for other studies.

  • implementation science
  • health services research
  • ambulatory care
  • chronic disease management

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information. Aggregate data generated or analysed during this study are included in this published article (and its supplementary information files). Raw data generated or analysed during this study are available from the corresponding author on reasonable request.

Statistics from Altmetric.com

Introduction

Cancer is considered a chronic illness, but cancer care lags in the integration of self-management support (SMS) into routine care. A recent global call to action recommended improvement in the quality of SMS in cancer1 given the evidence of benefit on health outcomes in chronic diseases2 and emerging evidence in cancer.3–5 Engagement of patients/survivors in self-management of cancer and health is central to quality cancer care6 and for achieving ‘Triple Aim’ healthcare goals of better health and patient experience at lower costs.7 Unfortunately, implementing SMS in routine care remains a challenge8 leaving patients with cancer vulnerable to worse health status and survival as they struggle to manage acute treatment toxicities and long-term effects.

Implementation science methods and evidence-based implementation strategies (eg, identify and prepare champions) are recommended for successful quality change.9 10 Both the knowledge-to-action framework11 and the Consolidated Framework for Implementation Research (CFIR)12 recommend tailoring implementation strategies to readiness and barriers/enablers within the local healthcare context.13–19 A taxonomy of evidence-based implementation strategies has been identified in the Expert Recommendations for Implementing Change (ERIC).20 However, there is little guidance on how to select and tailor implementation strategies to the healthcare setting in which the practice change is being implemented.21–24

Organisational readiness assessments (ORAs) are a diagnostic tool used to identify a range of contextual factors including barriers/enablers, characteristics of individuals, psychological and behavioural preparedness, implementation climate and innovation-specific factors that can impact on implementation.25–27 ORAs are crucial for tailoring an implementation approach to a care setting and as proximal outcomes towards achieving a distal outcome (ie, improved health), however their role in predicting effectiveness of implementation remains uncertain.27

There is a paucity of data on successful implementation approaches to facilitate integration of SMS in cancer care.28 Moreover, a systematic process for selecting and tailoring implementation strategies to organisational and clinician readiness, and barriers/enablers from empirically derived data is lacking.29–33 In this pre-implementation, formative phase of our multisite, pilot randomised controlled trial (RCT), Self-Management and Activation to Reduce Treatment-Related Toxicities (SMARTCare), we refine our implementation approach prior to testing a proactive model of SMS within three regional cancer centres (RCCs). The study aims were to assess organisational readiness and barriers/enablers for delivering SMS in routine care and map SMS readiness and barriers/enablers to CFIR constructs to derive implementation strategies to facilitate change.

Methods and procedures

Overview

We used a convergent parallel, mixed-methods design34 comprising validated surveys (quantitative data) and qualitative focus groups/interviews of key stakeholders (oncologists, nurses, allied health professionals and administrative leaders) at three participating RCCs in Ontario, Canada. Focus groups/individual interviews35 were used to obtain further insight into quantitative results and local context using a qualitative description methodology which focuses on who, what and where questions to gain insights about poorly defined phenomena.36 The SMARTCare RCT was approved through Clinical Trials Ontario, a centralised ethics review organisation for multicentre studies in Ontario and registered with clinicaltrials.gov. Written consent was obtained from participants electronically prior to survey completion and before audiotaping of focus groups and interviews.

Participants

Eligible participants (oncologists, nurses, social workers, pharmacists, dieticians) and RCC programme administrators and clinical leaders (ie, disease site group managers) from the colorectal, lymphoma and lung cancer disease site groups at the three RCCs were invited to participate via email. Other allied personnel were not approached as they often have variable contact with patients (ie, psychologists, physiotherapists), are not assigned to a disease site team or are not an available resource in some centres. RCCs were purposively chosen given their differing patient volumes, geography and nursing care delivery models. Centre 1 is located in a mixed rural/small urban setting, with mid-size patient volume and an integrated model of care, whereas centres 2 and 3 are located in large cities, have higher patient volumes and are academic cancer centres. Centres 1 and 2 follow a primary nursing model (nurses share responsibility for patient management with a specific medical oncologist), whereas centre 3 follows a team-based model of care whereby a team of nurses provide care to a disease site population and are not assigned to a specific oncologist.

Data collection

Quantitative survey data

Quantitative survey data were collected using the online Qualtrics survey platform (Qualtrics; Provo, Utah, USA) or via paper if participants were unable to complete online. The survey took approximately 20–30 min to complete and consisted of demographic questions and four validated measures. A modified Dillman approach37 was used to increase participation rates, whereby a local administrative or disease site lead within the participating RCC sent an initial invitation email detailing the study to eligible participants with a link to the survey followed by a second reminder sent 14 days later. A small token of appreciation (coffee card) was provided for those who completed surveys or focus group or interviews.

Focus groups and interviews

Disease site team members and key stakeholders were also invited to participate in on-site focus groups or a telephone interview by disease site group managers. A member of the research team experienced in qualitative methods (DH) facilitated the focus groups/interviews using a semi-structured interview guide. A demographic form was completed by participants prior to the start of focus group/interviews, which were audio recorded and contemporaneous field notes taken to document contextual information. Audio-recordings were transcribed verbatim into de-identified text.

Measures

We operationalised organisational readiness as barriers and enablers based on constructs in implementation theories, specifically the CFIR framework and12 Normalisation Process Theory (NPT),38 clinicians’ beliefs and knowledge about SMS (individual characteristics),39 the extent of organisational and patient support elements for SMS as defined by the Chronic Care Model40 and implementation climate (commitment and efficacy for implementing the change).41 Implementation theories help to frame readiness assessments and the potential barriers/enablers to be addressed in a preplanning formative phase prior to implementation.42 43 NPT considers the social context of practice and the ‘work’ required to embed a new practice into everyday care.44 45 Four valid and reliable survey measures and a semi-structured qualitative interview guide were used.

Organisational Readiness for Change Assessment (ORCA) 46 measures readiness regarding the value of the change/innovation, quality of the organisational and leadership context to support the practice change and organisational capacity to facilitate the change across eight subscales. Initially, we reviewed mapping of CFIR constructs to ORCA47 and core elements of SMS48 49 to inform our innovation-specific questions (SMS and ambulatory cancer care context) for use in focus groups/interviews (online supplemental file 1, measure selection).

Supplemental material

Primary Care Resources and Supports for Chronic Disease Self-Management (PCRS) 50 51: was designed to be applied across different illnesses for measuring the integration of essential elements of SMS in two areas, organisational structure and patient support. Each item in the PCRS (32 items) is rated on a 10-point scale within four quality levels. A score of 1 is the lowest and indicates SMS is not part of everyday care, a score of 2–4 indicates SMS is sporadic, not delivered as part of a systematic approach and core elements (ie, goal setting and action plans) necessary for activation of patients’ self-management are seldom used routinely, a score of 5–7 indicates essential elements of quality SMS are occurring but not fully integrated, and 8–10 indicates full integration of SMS in everyday care.

Clinician Support-Patient Activation Measure 52 (13 items, Likert scale 1–4) assesses clinician’s beliefs and attitudes about patient activation for illness self-management. Beliefs and attitudes are an individual characteristic as per CFIR that can influence uptake of SMS. Higher scores indicate clinicians value patient activation in self-management.

Organisational Readiness for Implementing Change (ORIC) 41 (10 items (Likert scale 1–5, two subscales), measures shared resolve to implement a change (commitment) and beliefs in collective capability to implement change (self-efficacy). A higher score indicates greater commitment and efficacy.

Semi-Structured Interview Guide: the semi-structured interview questions were based on CFIR and NPT constructs. NPT identifies how healthcare providers: (1) make sense of a new practice for uptake in everyday work (coherence); (2) what relational work teams do to redesign the work around a new practice and sustain its use (cognitive participation); (3) skill sets and interactions to make a new practice workable and to foster team accountability (collective action) and (4) how clinicians appraise the impact of a new practices on self and the team to inform work reconfiguration (reflexive monitoring).44 As an example, the following qualitative questions based on NPT were posed: “what changes in roles or responsibilities, or how the team works together do you anticipate to implement SMS in routine care” and “what strategies will be important to influence practice change, such as training needs, knowledge and skills to create a common understanding of SMS”.

Quantitative data analysis

Descriptive statistics were used to summarise participant characteristics. Means, median and SD for each question, and subscales scores were calculated across all respondents for survey measures and stratified by RCC.

Classifying barriers and enablers from quantitative data using CFIR

Median scores were classified as ‘barriers’, ‘enablers’ or ‘neutral’ (no effect on implementing SMS) based on a priori measure-specific threshold definitions based on consensus among the study team members: PCRS (enabler: 8–10; neutral: 5–7; barrier: 1–4), CS-PAM (enabler: 4; neutral: 2–3; barrier: 1), ORCA (enabler: 5; neutral: 2–4; barrier: 1) and ORIC (enabler: 5; neutral: 2–4; barrier: 1). High intra-RCC variability was believed to indicate there was potential for additional implementation issues at the site due to lack of consensus. As such, questions with high variability in responses were also coded as potential barriers. The SD was chosen as >2 for the PCRS and >1 for the CS-PAM, ORCA and ORIC, as the measures are based on 10-point and 5-point scales, respectively. Three team members individually coded the barriers and enablers generated from each survey measure to the CFIR domains and constructs and generated a consensus.

Identifying implementation strategies using the CFIR-ERIC Strategy Matching Tool V.1.0

Identified barriers and enablers from surveys were then individually mapped to the CFIR domains and constructs. Barriers were then inputted into the CFIR-ERIC Strategy Matching Tool V.1.0 (CFIR Research Team; Ann Arbor, Michigan, USA) to generate a list of implementation strategies tailored to context-specific barriers by RCC. Implementation strategies with a cumulative match of 50% or greater were identified as relevant. Discrete ERIC implementation strategies described by Powell et al 20 were used to organise the CFIR-ERIC strategies based on strategy and subgroup, and included in the analysis.

Qualitative data

Transcripts from focus groups/interviews were analysed by two qualitative research experts (DH, HA) using a reflexive thematic analysis approach with initial deductive framework coding based on domains and constructs from CFIR followed by inductive coding to derive themes grounded in the data.53 54 Text segments were given content-specific codes in NVivo V.10 (QSR International; Doncaster, Victoria, Australia) and emerging themes, barriers/enablers and implementation tactics identified for each RCC. Confirmability55 was addressed through independent coding of select transcripts followed by discussions to reach consensus on a coding framework and verified with other research team members, which was applied to all data.56 Credibility of data was enhanced through inclusion of diverse stakeholders.57 Quantitative and qualitative data were analysed separately and then triangulated using a convergence-coding matrix58 to obtain a deeper understanding of barriers/enablers unique to each disease site group and RCC. Site-specific, tailored implementation recommendations were generated based on the barriers/enablers, and identified strategies fed back to RCCs to guide development of tailored implementation plans.

Results

Participant characteristics

Of the 157 stakeholders approached, 84 agreed to participate in the web-based survey. Of the 84 stakeholders surveyed, 78 (respondent rate=50%; range: 33.3%–58.9%) completed the survey and were included in the final analysis (online supplemental file 2, consort chart). Among all respondents, 57.1% worked in colorectal cancer, 56% in lung cancer and 50% in lymphoma (table 1). Age varied across participants and most had been working in oncology for 6 or more years (61.9%), identified as a nurse (47.6%) or medical oncologist or haematologist (32.1%), and were permanent full-time employees (75%).

Supplemental material

Table 1

Participant demographic characteristics from survey data by centre

Quantitative survey results

Median, mean and SD were calculated for each item in the survey measures (online supplemental file 3); mean subscale scores are shown in table 2. While overall mean scores on the PCRS did not vary much between sites, the scores indicated that none of the RCCs had fully integrated SMS. Centre 1 had higher integration of SMS overall. In terms of the PCRS, scores were low for patient support elements (mean range: 4.76–5.90) across all centres, specifically for SMS techniques of goal setting/action planning and problem-solving. Scores were lowest for centre 3 (mean: 4.6) compared with centre 2 (mean: 5.33) and centre 3 (mean: 5.90). For organisational support elements, scores were lowest for centre 3 (mean: 4.53) compared with centre 2 (mean: 5.12) and centre 1 (mean: 6.44). Clinicians in all centres had high scores on the CS-PAM (mean range: 3.65–3.74) indicating that they valued patient activation in self-management. Organisational readiness (ORCA) means ranged from 3.18 to 4.67 across sites. Scores were lowest for centre 3, particularly for items of resources and leadership practice. ORIC was highest for centre 1 for both change efficacy (mean: 3.51) and commitment (mean: 3.52).

Supplemental material

Table 2

Means and SD for subscale scores for quantitative survey measures

Mapping survey data to CFIR domains and constructs

Using the measure-specific threshold definitions, a total of 9 barriers and 27 enablers were identified for centre 1, 23 barriers and 14 enablers for centre 2 and 23 barriers and 13 enablers for centre 3 (table 3). Once mapped to the CFIR domains and constructs, 7 barriers and 12 enablers were identified for centre 1, 14 barriers and 9 enablers for centre 2 and 11 barriers and 5 enablers for centre 3. There were five CFIR enablers common to all three sites: inner setting—compatibility and characteristics of individuals—knowledge and beliefs about the intervention and self-efficacy. Six CFIR barriers were common to all three sites: inner setting—networks and communication, goals and feedback, readiness for implementation and access to knowledge and information. CFIR domains and constructs identified as both a barrier and an enabler for all three sites were: inner setting—implementation climate and characteristics of individuals—other personal attributes.

Table 3

CFIR domains and constructs and associated enablers and barriers by centre

Focus groups and interviews

Twenty-four disease site team members participated in focus groups/interviews (centre 1, n=9; centre 2, n=5; centre 3: n=10). Most were female (83.3%), had a bachelor’s degree or higher (75.1%), had been practising in oncology for 6 or more years (75%), were a nurse (45.8%) or medical oncologist/haematologist (37.5%) (online supplemental file 4). Among all respondents, 33.3% worked in colorectal cancer, 50% in lung cancer and 54.2% in lymphoma.

Supplemental material

Theme 1: SMS is highly valued by all team members. Consistent with scores on the CS-PAM, all sites valued SMS and thought it would reduce ‘symptom crises’. However, they perceived a lack of ‘coherent’ understanding of SMS across team members and a need for clear designation of team roles/responsibilities and expectations for performance (ie, reduced urgent calls). While they saw the potential ‘workability’ of SMS in their practice they did not want to disrupt current workflow; and were uncertain how this could be implemented given time constraints, complexity of patient needs and inadequate ‘funding for frontline care’.

Theme 2: Shift paternalistic culture to engage patients as partners in care. Participants viewed the current paternalistic culture as a key barrier. They described the culture as ‘supportive’ and of ‘carrying patients through treatment’ rather than enabling them to become effective self-managers. They voiced a need for patients to ‘take responsibility for managing their symptoms’ and that perhaps a ‘written contract between patients and clinicians’ would be necessary.

Theme 3: SMS tools and skills will enhance existing nursing practice. Across the three sites, participants felt they were providing SMS, but their main tactic to achieve this was to educate patients about all treatment side effects to ‘keep them safe’ and were uncertain if patients acted on this information. The need to systematise delivery of SMS and skills of all team members through targeted training and communication for ‘buy-in’ and to reconfigure educational processes to integrate SMS with journey mapping was identified.

Qualitative data revealed subtle variation in readiness to implement SMS and barriers/enablers. Centre 1 was receptive to implementing and learning about SMS, expressed strong leadership support and a climate of quality improvement. Whereas centre 2 was more suspicious of change, expressed concerns about leadership support and indicated there was a lack of champion authenticity and engagement of managers in other practice innovations that failed. Centre 3 expressed a strong desire to have a ‘centralised area for SMS resources’, thought SMS would be seen as an ‘extra’ by oncologists and were least familiar with SMS in clinical care. Ongoing communication and real-time feedback to sites about the impact of SMS (ie, less urgent calls from patients in symptom crises) were considered key to quality monitoring.

ERIC implementation strategies

Using the CFIR-ERIC Strategy Matching Tool V.1.0, 30 implementation strategies were identified for centre 1, 46 for centre 2 and 41 for centre 3. A total of 46 overarching implementation strategies were identified, of which 30 (65%) were common to all RCCs. The ERIC strategy with the highest cumulative percentage match for centre 1 was ‘identifying and preparing champions’ and for centre 2 and 3 it was ‘assessing for readiness and identifying barriers and facilitators’ (data not shown). Results for centre 2 and their perspective that few pilot projects had translated into practice change and their ‘suspicions’ regarding success of change and selection of champions suggested that external and internal practice change facilitators would be needed. Following mapping of quantitative barriers from survey data to CFIR constructs and CFIR-ERIC matching to identify implementation strategies, we mapped the qualitative data to CFIR barriers and enablers (online supplemental file 5(a,b)) and to ERIC implementation strategies and identified specific tactics (online supplemental file 6).

Supplemental material

Supplemental material

Site-specific implementation recommendations

Implementation strategies with corresponding subrecommendations were generated for all three centres based on the ERIC matching process and triangulation of quantitative and qualitative data sources (figure 1). The implementation recommendation unique to centre 2 was ‘identify and engage appropriate champions and leadership’ and unique to centre 3 was ‘create a centralised area with resources’. Both centre 2 and 3 identified the need for quality monitoring tools and learning from others through site visits.

Figure 1

Tailoring of implementation strategies based on ERIC and mapping to quantitative and qualitative data sources. SMS, self-management support.

Discussion

There is increasing evidence that activation of patients in self-management is essential to optimise health outcomes in chronic diseases.1 A key issue to be addressed is how to implement and integrate SMS in the context of rapid, episodic ambulatory care with large patient volumes and care complexity. Our readiness survey data revealed that oncology clinicians hold positive beliefs about patient activation in self-management, however some differences were noted across RCCs regarding the relative advantage, barriers/enablers and leadership support for implementing this change. Similar to other research,8 clinicians had little understanding of the core elements of SMS and assumed patient education was SMS. Goal-setting/action planning59 and other organisational elements fundamental to effective SMS were not integrated into routine care. Thus, ‘symptom crises’ were reported as common, which is likely reflected in the high rates of emergency department use (>40%) during systemic chemotherapy in Ontario.60

Our use of measure-specific threshold definitions for the four quantitative readiness surveys enabled us to organise the baseline data into clearly defined barrier/enabler/neutral categories, possibly reducing the implicit bias in current approaches. Ross et al 33 used CFIR constructs to identify contextual factors and inform selection of implementation strategies based on a literature review. Lewis et al 61 used the Damschroder classification of CFIR constructs to organise barrier/enablers and generate implementation strategies using a methodology similar to ours. However, rather than using the CFIR-ERIC, Lewis et al 61 ranked the Powell et al 20 discrete implementation strategies based on stakeholders’ perceptions of feasibility and importance, and self-selected implementation experts refined the matching to barriers and implementation phases. The CFIR domains and constructs provided a useful framework for identifying barriers/enablers to be addressed in implementation.62 63 NPT enabled deeper insight into the work of embedding SMS in routine care, although the NPT constructs have not yet been specifically mapped to ERIC strategies. While evidence-based ERIC implementation strategies are key to uptake of innovations,20 they are generic and further innovation-specific tailoring is required for implementation success.27 Combining established frameworks and quantitative survey findings with context-specific qualitative interviews enabled a deeper understanding of tailoring specific for SMS in cancer care.

While data were gathered from three diverse disease sites and RCCs increasing the relevance of our findings, our findings may not be generalisable to other jurisdictions where care may be organised differently. Our survey response rates were low (50%) and mainly comprised oncologists and nurses and not the broader range of multidisciplinary stakeholders involved in cancer care. Our research team mainly comprised physicians and nurses possibly introducing disciplinary bias. Stakeholders who chose to participate in the focus groups/interviews might also have had a greater interest in the study than non-participants potentially introducing self-selection bias.

Our readiness data showed clinician support for SMS but a need for system redesign including changes to staffing, workflow and finances to systematise SMS into routine cancer care. Addressing the barriers to integrating SMS to enable patients more control over their care may be particularly challenging in highly routinised settings such as ambulatory cancer care. Other researchers have noted that underestimating the complexity of effective SMS implementation in routine care can lead to implementation failure and identified the need for a ‘whole system’ change approach.64

Although there is little empirical evidence of SMS implementation in routine cancer care, evidence from ‘whole’ system disease self-management change approaches in the UK65 and primary care suggest a need to optimise staff roles/responsibilities, and for upskilling of clinicians’ skills in person-centred communication and use of practical evidence-based, behaviour change counselling techniques such as the five As (5As) (ask, advice, assess, assist, arrange).66–68 Effectiveness of the 5As has been shown for different behaviours (ie, physical activity, smoking cessation, obesity) and health conditions (ie, diabetes), and is feasible in busy practice settings.69–71 As recommended by our participants, centralised written or video resources to learn about SMS may be necessary.

Participants identified the need to integrate SMS into workflow to avoid it being seen as an ‘optional extra’ to existing care processes. Strategies such as pre-preparation of patients (eg, online assessment of self-management capacity), capitalising on wait times, group visits, technology or community partnerships to deliver SMS are suggested.72 Existing patient education workflow could also be enhanced to incorporate simple SMS techniques such as goal setting and action plans, which are shown to reduce emergency department visits in other chronic diseases.2

Better financing of front line clinic staff and financial incentives were identified by our participants as implementation strategies in ERIC taxonomy mapping.20 Financial incentives such as pay-for-performance schemes have been proposed to improve the quality of healthcare services in chronic diseases with effects on process outcomes (ie, monitoring blood glucose in diabetes),73 but evidence of different approaches is limited.74 Financial incentives for changing healthcare professionals’ behaviour has shown mixed-effectiveness and further robust research is required.75

Conclusion

In spite of quality standards on SMS developed for RCCs in Ontario, Canada,76 our data show that the quality of SMS still requires site-specific improvement. For implementation science to move forward, methodologies for tailoring implementation strategies needs further refinement and validation. Our systematic process may have relevance for other studies, however the practicality of using multipronged implementation strategies in resource-constrained environments requires evaluation and to determine the key strategies for uptake of SMS in routine practice. Multipronged actions are recommended for integrating SMS into routine cancer care such as performance metrics and training of healthcare professions and patients.1 Future research to evaluate effectiveness of SMS in cancer populations and implementation in ‘real-world’ cancer settings are identified as a priority.28

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information. Aggregate data generated or analysed during this study are included in this published article (and its supplementary information files). Raw data generated or analysed during this study are available from the corresponding author on reasonable request.

Ethics statements

Patient consent for publication

Ethics approval

The study has been approved through Clinical Trials Ontario (CTO), a centralised ethics review organisation for multicentre studies that is used by the three participating cancer centres. Approval number: 1371.

References

Supplementary materials

Footnotes

  • Contributors Planning: DH, MP, RK, LM, MAO'B, MK. Conduct: DH, MP, RK, HA, LM, DB-L, MAO'B, SR, MK. Reporting: DH, MP, RK, HA, LM, DB-L, MAO'B, SR, MK. Overall content as guarantor(s): DH.

  • Funding Funding for this project was provided by the Canadian Institutes of Health Research’s (CIHR) Operating Grant (HRC 154129): Partnerships for Health System Improvement for Cancer Control, and in-kind contributions by Cancer Care Ontario.

  • Disclaimer The opinions, results and conclusions reported in this paper are those of the authors and are independent from the funding sources. No endorsements by CIHR or Cancer Care Ontario is intended or should be inferred.

  • Competing interests HA and LM were employees of Cancer Care Ontario (CCO, now part of Ontario Health), who provided in-kind support for this study.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles