Sage Journals: Discover world-class research

Abstract

Mixed methods research offers a unique opportunity to advance understanding of a phenomenon. However, practical guidance on the use of mixed methods to develop theoretical frameworks is limited. We present a novel adaptation of the Pillar Integration Process (PIP) to support development and refinement of a theoretical framework explaining the processes by which healthcare factors influence cancer survivors’ quality of life. Deductive integration of quantitative and qualitative data, guided by an a priori conceptual model, illuminates the nuances of the relationship between the concepts of interest, producing a theoretical framework, the Model of Healthcare factors influencing Quality of Life in Cancer Survivorship (MoHaQ-CS).

Keywords

cancer survivorship Pillar Integration Process mixed methods quality of life theoretical framework

Introduction

Mixed methods research (MMR) has an advantage over quantitative or qualitative research in its potential to produce knowledge that is unavailable in either method alone (Creswell and Clark, 2017). However, there is often a failure to adequately integrate the findings of MMR to realize their full potential (O'Cathain et al., 2007). The products of integration are the meta-inferences drawn from combining quantitative and qualitative results, yielding a more complete understanding of a research problem (Creswell and Clark, 2017). However, in sequential MMR studies, quantitative and qualitative data analysis are often undertaken and presented sequentially, potentially compromising the development of meta-inferences (Bazeley, 2018).

Theoretical frameworks are essential drivers of empirical research, linking concepts and variables to understand complex phenomena, based on established and emerging evidence (Hammer et al., 2019). Theoretical frameworks provide schemas to situate the research problem, guide the selection of data collection instruments and support the development of complex health interventions and services (Hammer et al., 2019). However, discussion of the theoretical bases of research is often not sufficiently visible in the development, implementation, evaluation or reporting of complex interventions, limiting evaluation and evidence synthesis (Cuthbert et al., 2019). Underutilization of theoretical frameworks in the field of implementation science has been attributed to users’ knowledge, confidence, skills and experience with their use (Lynch et al., 2018).

Development and refinement of theoretical models are traditionally based on literature reviews and qualitative methodologies, with testing and validation via predominantly quantitative methods (Kivunja, 2018). However, MMR offers a unique opportunity to comprehensively evaluate and refine theory to advance understanding of a phenomenon, representing a high standard of integration in MMR. However, practical guidance in the use of MMR in the advancement or development of theoretical frameworks is limited (Evans et al., 2011).

This paper presents the adaptation of the Pillar Integration Process (PIP) (Johnson et al., 2019) to support the development and refinement of a theoretical framework to explain the processes by which healthcare factors may influence colorectal cancer (CRC) survivors’ QoL, the Model of Healthcare factors influencing Quality of Life in Cancer Survivorship (MoHaQ-CS). PIP advocates for an inductive approach to analysis; in the first stage, raw data are listed in separate quantitative and qualitative columns. In the second stage, the data is matched through a process of coding and categorizing similar qualitative and quantitative data in a joint display. In the third stage, the data and categories of the joint display are checked for completeness, ensuring data are appropriately matched, and no raw data could provide an appropriate match for identified gaps in the joint display. In the fourth and final stage, Pillar Building is conducted through comparison and contrast of findings developed through the listing, matching and checking of data. Here, we present an alternative, deductive approach to PIP, guided by the Ashing-Giwa (2005) Contextual Model of Health-Related Quality of Life.

Study Background

A primary goal of healthcare is to optimize an individuals’ QoL. Follow-up care is an integral part of life after cancer treatment, to identify symptoms of potential cancer recurrence and support recovery and management of long-term side-effects (Institute of Medicine, 2006). Despite the importance of follow-up care for physical and psychosocial recovery, there is little understanding of how the experience of healthcare may affect cancer survivors’ well-being, including access, continuity, outcomes of and satisfaction with care (Drury et al., 2017; Sisler et al., 2012).

CRC survivors experience a range of distressing physical, psychological and social issues which may impact their well-being for many years after cancer treatments (Drury et al., 2017, 2021). Cancer survivors require high-quality survivorship care to support recovery and adjustment to chronic physical, psychological and social issues after cancer. However, the organization of healthcare services tasked with survivorship care are complex, encompassing many variables, involving multiple healthcare providers, in often fragmented care pathways spanning several organizations (Gordon et al., 2012; Sisler et al., 2012; Snyder et al., 2008). Furthermore, surveillance for cancer recurrence may be prioritized above the management of survivorship issues in some circumstances (Di Fabio et al., 2008; Tofthagen, 2010). As a result, cancer survivors may experience a range of unmet information and supportive care needs affecting their QoL (Cheng et al., 2016; Hallett et al., 2014).

QoL and perceptions of well-being are unique to each person, influenced by individual values, expectations and culture (World Health Organisation Quality of Life Group, 1995). While healthcare is recognized as a determining factor in cancer survivors’ QoL outcomes, few QoL models explicitly acknowledge the broader healthcare context. The Ashing-Giwa (2005) Contextual Model of Health-Related Quality of Life has been developed and tested with ethnically and culturally diverse samples of breast and cervical cancer survivors. However, the models’ focus on and basis in cancer survivorship and the recognition of healthcare as influencing QoL outcomes, alongside demographic, socio-ecological and cultural factors means it provides a comprehensive framework to explore factors placing cancer patients and survivors at greater risk of experiencing poorer QoL (Ashing-Giwa and Lim., 2008, 2011). To date, few studies have attempted to advance understanding of the interaction between healthcare experiences and QoL outcomes. To ensure complex interventions are responsive to the supportive care needs and QoL of cancer survivors, a comprehensive theoretical framework is required. We present the MoHaQ-CS, which describes the processes by which healthcare systems may influence QoL outcomes among cancer survivors.

Methods

The MoHaQ-CS was developed based on the integration of quantitative and qualitative phases of The Cost of Survival Study, a pragmatic mixed methods sequential explanatory study exploring the healthcare factors associated with CRC survivors’ QoL (Figure 1). The design and methods of each phase of the study are reported elsewhere (Drury et al., 2020a, 2020b, 2021). Table 1 summarizes the aims, sampling and data collection methods utilized within each phase of the study. Participants of the study were CRC survivors of any disease stage, between 6 and 60 months post-diagnosis, aged 18 years old or older and actively engaged in cancer surveillance and follow-up care in Ireland. 304 CRC survivors were recruited to the quantitative phase from public and private hospitals and voluntary cancer support centres (response rate = 75.6%). For the qualitative phase, a subsample of 22 consenting survey participants (n = 238, 92.2%) were invited to participate in semi-structured interviews using a maximum variation sampling strategy to ensure representation of key demographic, health and healthcare-related variables associated with CRC survivors’ QoL outcomes in the quantitative phase of the study (Table 1). During the quantitative phase, participants completed subjective measures of QoL with established validity and reliability with English-speaking samples of CRC survivors (Table 1) (Sisler et al., 2012; Ward et al., 1999; Wright et al., 2011). Semi-structured telephone interviews were conducted with CRC survivors to explore their experiences of cancer survivorship and the perceived impact of cancer-related care upon their QoL.

Figure 1.

The Cost of Survival Study Mixed Methods Sequential Explanatory Study Design.

Table 1.

The Cost of Survival Study Research Methods.

Phase	Aims	Participants	Data collection
Phase 1: Quantitative	To understand CRC survivors’ physical, psychological and social outcomes. To determine the healthcare-related factors which influence CRC survivors’ QoL outcomes.	304 CRC survivors: • between 6 and 60 months post-diagnosis, • aged 18 years old or older, • engaged in cancer surveillance and follow-up care	Questionnaire • EuroQOL • Functional Assessment of Chronic Illness Therapy–Colorectal (FACT-C) questionnaire (Ward et al., 1999) • Social Difficulties Inventory (SDI) (Wright et al., 2011) • Patient Continuity of Care Questionnaire (PCCQ) (Hadjistavropoulos et al., 2008)
Phase 1: Quantitative			Individual items evaluating perceptions and experiences of cancer care derived from: • National Cancer Patient Experience Survey (Department of Health United Kingdom 2010) • Patient Reported Outcome Measures Living with and Beyond Colorectal/Gastro-Intestinal Cancer Questionnaire (Quality Health, 2012) Analysis: Descriptive, inferential and logistic regression analysis.
Phase 2: Qualitative	To explore the impact of physical, psychological and social effects upon CRC survivors’ QoL. To provide a contextual understanding of the healthcare experiences and unmet needs reported by CRC survivors in the quantitative study.	Subsample of 22 Phase 1 participants. Maximum variation sampling strategy representing variations in: • Time since diagnosis • Cancer status • Type of healthcare • Living arrangements • FACT-C score • Self-rated health • Gender • Age	Semi-structured interview, exploring participants’ QoL and symptom experience, experience of supportive care and information needs. Central Interview Questions: • Could you please tell me about your experiences of living with/after colon/rectal cancer? • Could you tell me about the healthcare you have received since your treatment for colon/rectal cancer? Analysis: Inductive thematic analysis (Braun et al., 2006).

A strong emphasis was placed on achieving multiple levels of integration in the planning, design and analysis stages of this study (Table 2). The Cost of Survival Study received ethical approval from the Research Ethics Committees of the School of Nursing and Midwifery, Trinity College Dublin (Reference: The Cost of Survival) and participating hospitals (Reference: 2014/05 Chairman’s Action (6); Reference: 06/2014 The Cost of Survival). Methodological rigour in this study was ensured through the use of strategies to address the standards of quality appraisal for MMR; veracity, consistency, applicability and neutrality (Curry, 2015) (Table 3).

Table 2.

Points of Integration in the Mixed Methods Cost of Survival Study.

Points of integration in the Cost of Survival Study
• Integrated research objectives derived from gaps in the literature and guided by an a priori theoretical framework.
• Quantitative findings informed the development of the qualitative interview schedule.
• Use of an integrated nested sampling strategy, inviting a subsample of survey participants to take part in follow-up interviews.
• Quantitative results informed the development of the maximum variation sampling strategy used to identify interview participants of the qualitative phase.
• Integration of findings responding to the mixed methods objectives is presented within this manuscript, identifying convergence and divergence of results between quantitative and qualitative phases.
• The integrated results of this study inform the development of a theoretical framework presented within this paper, which demonstrates the impact of healthcare, socio-ecological and cultural contexts of the Ashing-Giwa (2005) Contextual Model of Health-Related Quality of Life upon colorectal cancer survivors’ quality of life.
• Meta-inferences drawn from the integration of quantitative and qualitative findings are summarized and discussed within this paper.

Table 3.

Application of Mixed Methods Quality Standards in the Cost of Survival Study (Curry, 2015).

Standard (Curry, 2015)	Quantitative criteria	Qualitative criteria
Veracity	Internal Validity	Credibility
The degree to which results accurately and precisely represent the phenomenon under study.	Advisory panel appointed to establish face validity of quantitative measures.	Prolonged engagement building trust between the researcher and participant.
		Collection of field notes, audiotaped interviews, verbatim and member checking of transcripts.
		Random selection of participants meeting the maximum variation sampling criteria.
		Ongoing critical evaluation of data collection activities and analytical processes through reflexive journaling and debriefing with research team.
Consistency	Reliability	Dependability
Underpinned by the reliability of quantitative research and the dependability of qualitative research.	The study used questionnaires with established validity and reliability.	Consideration of participant and research perspectives during interpretation of qualitative analysis through reflexive journaling.
	Each instrument was assessed for internal consistency and achieved acceptable levels of α≥.7	Systematic audit trail, detailing analytical decisions using the coding, memo and annotation functions of NVivo.
Applicability	External Validity	Transferability
The potential for study results to translate to a different settings or populations.	Participants of the current study are drawn from several sites with different designations (centre of excellence vs. regional vs. private), varying sources of funding (public vs. private healthcare) and diverse approaches to healthcare (nurse-led vs. consultant-led), enhancing the generalizability of results.	Rich, detailed descriptions of the study context, time, place and culture provided.
Neutrality	Objectivity	Confirmability
Cognizance of the assumptions which may bias the implementation of a study or the interpretation of results.	Transparency in the reporting of quantitative methods and analytic decisions	Presentation of participant quotations to support analysis and interpretation of qualitative data.
		Systematic audit trail, detailing analytical decisions using the coding, memo and annotation functions of NVivo.
		Reflexivity on the researchers’ influence on the research process

Data Analysis and Integration

The approach to data analysis and integration within this study was iterative, whereby quantitative and qualitative data were analysed sequentially, and subsequent integrated analysis was guided by a modified, deductive PIP (Figure 2) (Johnson et al., 2019). Stages 1–3 of the 4-stage PIP were conducted deductively, whereby the domains of the Contextual Model of Health-Related Quality of Life provided the pillars of analysis. In the first stage of the modified PIP, raw quantitative and qualitative data and key findings were listed. In stage 2, the contents of the quantitative and qualitative lists were matched to the pillar domains of the Ashing-Giwa (2005) model, and in stage 3, they were checked for completeness and appropriate matching. In the fourth and final stage, Pillar Building, the joint display was analysed to formulate meta-inferences which provide a more comprehensive understanding, explanation and context of the healthcare factors influencing CRC survivors’ QoL. This analysis enabled the generation of a visual illustration of the MoHaQ-CS, demonstrating the processes by which cancer-related healthcare factors may influence cancer survivors’ QoL, moderated by individual and macro/systemic domains of the Ashing-Giwa (2005) Model.

Figure 2.

Modified Pillar Integration Process adapted for the Cost of Survival Study Study.

Results

Sample Characteristics

The demographic characteristics of participants of the quantitative and qualitative phases of The Cost of Survival Study are presented in Table 4.

Table 4.

Sample Characteristics.

Characteristic	Response	Quantitative sample		Qualitative sample
Characteristic	Response	n	%	n	%
Demographic Context
Age	<65	93	32.9	10	45.5
Age	≥65	190	67.1	12	54.5
Gender	Female	126	44.2	10	45.5
Gender	Male	159	55.8	12	54.5
Socio-Ecological Context
Living Arrangements	Lives with Others	231	81.3	18	81.8
Living Arrangements	Lives Alone	53	18.7	4	18.2
Area of Residence	Urban	215	76.5	12	54.5
Area of Residence	Rural	66	23.5	10	45.5
Change in Employment Status Since Diagnosis	Remained/Became Employed	71	25.0	9	40.9
	Remained Unemployed	167	58.8	9	40.9
	Became Unemployed	46	16.2	4	18.2
Private Health Insurance	Yes	138	48.3	13	59.1
Private Health Insurance	No	148	51.7	9	40.9
Social Difficulties	None	96	34.0	3	13.6
Social Difficulties	One or More	186	66.0	19	86.4
Healthcare Context
Type of Hospital Attended	Centre of Excellence	164	56.9	9	40.9
	Private	24	8.3	5	22.7
	Regional/Other	100	34.7	8	36.4
Level of Satisfaction with Continuity of Care	Satisfied	161	59.6	12	54.5
Level of Satisfaction with Continuity of Care	Some Dissatisfaction	109	40.4	10	45.5
Care Plan	Yes	76	26.9	5	22.7
Care Plan	No	207	73.1	17	77.3
Treatment Summary	Yes	97	34.6	4	18.2
Treatment Summary	No	183	65.4	18	81.8
Access to a Named Nurse for Cancer-Related Worries	Yes	227	80.8	14	66.7
Access to a Named Nurse for Cancer-Related Worries	No	54	19.2	7	33.3
Access to a Named Doctor for Cancer-Related Worries	Yes	226	79.9	15	68.2
Access to a Named Doctor for Cancer-Related Worries	No	57	20.1	7	31.8
Feel supported by Hospital Staff	Yes	269	95.1	20	90.9
Feel supported by Hospital Staff	No	14	4.9	2	9.1
Feel Supported by Primary Care Professional	Yes	215	76.2	17	77.3
Feel Supported by Primary Care Professional	No	67	23.8	5	22.7
Feel Supported by Community Health Staff	Yes	171	61.1	12	54.5
Feel Supported by Community Health Staff	No	109	38.9	10	45.5
Unmet Information Needs	None	88	32.4	7	31.8
Unmet Information Needs	One or More	184	67.6	15	68.2
Cultural Context
Ethnicity	Irish	274	95.5	22	100.0
Ethnicity	Other	13	4.5	0	0.0
Accessed Voluntary Cancer Support Services	No	239	85.1	18	81.8
Accessed Voluntary Cancer Support Services	Yes	42	14.9	4	18.2
Cancer-Specific Medical Factors
Diagnosis	Colon	191	64.1	13	59.1
	Rectum	69	23.2	7	31.8
	Other	38	12.8	2	9.1
Time Since Diagnosis	<2 years	110	39.6	7	31.8
Time Since Diagnosis	≥2 years	168	60.4	15	68.2
Radiotherapy	None	228	76.0	12	54.5
Radiotherapy	Any	72	24.0	10	45.5
Chemotherapy	None	125	41.7	3	13.6
Chemotherapy	Any	175	58.3	19	86.4
Surgery	None	26	8.7	2	9.1
Surgery	Any	273	91.3	20	90.9
Stoma	Never had a Stoma	153	53.5	10	45.5
	Stoma Reversed	78	27.3	7	31.8
	Stoma Present	55	19.2	5	22.7
General Health and Morbidity
Disease Status	In Remission	228	82.9	20	90.9
Disease Status	Any Active Disease	47	17.1	2	9.1
Comorbidities	None	62	22.1	4	18.2
Comorbidities	One or More	218	77.9	18	81.8

Integrated Findings

The quantitative and qualitative findings each aligned with the core domains of the Ashing-Giwa (2005) Contextual Model of Health-Related Quality of Life during initial sequential analysis of each dataset. The regression models developed in the quantitative phase fit with Ashing-Giwa’s conceptualization of QoL, highlighting the importance of demographic, socio-ecological, healthcare and cultural contexts in predicting CRC survivors’ QoL (Figure 3). The qualitative findings support and extend these results, suggesting that formative healthcare experiences, particularly failures in relational, informational and management aspects of continuity of care during diagnosis and treatment may continue to adversely affect QoL, even 5 years after diagnosis (Figure 4). Quantitative and qualitative data were subsequently integrated following the modified PIP process (Table 5), which are thematically summarized below.

Figure 3.

Quantitative factors associated with colorectal cancer survivors’ quality of life outcomes, grouped according to the domains of the Ashing-Giwa (2005) Contextual Model of Health-Related Quality of Life.

Figure 4.

Conceptual map of qualitative of factors influencing colorectal cancer survivors’ quality of life outcomes, grouped according to the domains of the Ashing-Giwa (2005) Contextual Model of Health-Related Quality of Life.

Table 5.

Pillar Integration Process Joint Display: Integrated Results of the Cost of Survival Study.

Quantitative key findings (Drury et al., 2020b, 2021)	Quantitative data	Pillar (Ashing-Giwa, 2005)	Qualitative data	Qualitative key findings (Drury et al., 2020a, 2021)
Age and gender were associated with QoL for emotional, social and CRC-specific QoL in univariate and/or multivariate analysis.	Demographic factors associated with poorer outcomes for one or more QoL domains (p ≤ .05): • Age • Gender	Demographic Characteristics	Subtheme: Supportive Care, Does One Size Fit All? ‘Most people with colon cancer are a lot older than me … there was certainly a failure of experience and infrastructure and support groups, for somebody specifically of my age’. [RSM043]	This subtheme describes the experiences of younger people diagnosed with CRC, and how care was/was not tailored to meet their age-specific needs.
Employment status, insurance status, living arrangements and social difficulties, including financial problems, were associated with QoL for physical, social, emotional, functional, CRC concerns or overall QoL in univariate and/or multivariate analysis.	Socio-ecological factors associated with poorer outcomes for one or more QoL domains (p ≤ .05): • Employment Status • Insurance Status • Living Alone • Social Difficulties 25.4% (n = 71) reported financial difficulties on the Social Difficulties Inventory, which was associated with poorer QoL outcomes (OR = 5.444, CI: 2.854–10.386; p ≤ .001). 48.3% (n = 138) held private health insurance.	Socio-Ecological Factors	Subtheme: Living with the Impact of CRC: The financial challenges of cancer ‘[Social welfare] is €188 whether you like it or not and then you’ve to do the juggling’. [RSM027] ‘The financial help available from the state is also inadequate … the only reason I coped financially was from the help of my parents’. [ONL002] ‘I was coming home [from work], I would have my dinner, I’d go to bed, and I’d drag myself out of the bed in the morning, so I just couldn’t hack it’. [ESM036]	This subtheme describes the implications of cancer for cancer survivors’ employment and financial well-being.
Comorbidities were associated with QoL for physical, social, emotional, functional, CRC-specific and overall QoL in univariate and/or multivariate analysis.	General health and morbidity factors associated with poorer outcomes for one or more QoL domains (p ≤ .05): • Comorbidities • Disease status 77.9% (n=218) reported one or more co-morbid diseases.	General Health Factors	Subtheme: Striving to Regain, Maintain and Reconceptualize Control ‘I find I can’t take things in as well as before. It takes me longer if I’ve to read a complicated report now … I’m a bit more forgetful, but sure that’s probably age’. [ESM005] ‘Well there is, yes, a slight, a loss of energy but a certain amount of that is down to age too, I’m going to be seventy now in two months, so you have to accept that you’re not as fit as you were when you were twenty-five’. [ESM006]	This subtheme describes the strategies implemented by CRC survivors to self-manage and reconceptualize the physical consequences of living after CRC. Several survivorship issues were attributed to ageing and comorbidity, rather than cancer and cancer treatment.
Diagnosis, disease status, previous treatment with chemotherapy, and having a stoma were associated with QoL for physical, emotional, functional, CRC-specific and overall QoL in univariate and/or multivariate analysis. While physical, psychological and social survivorship issues were each associated with poorer QoL, less prevalent survivorship issues were associated with the highest likelihood of reporting poorer QoL.	Cancer-specific medical factors associated with poorer outcomes for one or more QoL domains (p ≤ .05): • Diagnosis • Chemotherapy • Stoma Negative body image (74%), fatigue (68%), sexual dysfunction (66%) and sleep disturbance (59%) were the most common late effects reported by CRC survivors, and each were associated with poorer QoL outcomes (p ≤ .05). Feeling ill (12.3%), family members’ difficulty accepting their loved ones’ diagnosis (20.0%), dissatisfaction with QoL (54%), difficulties with transport (14.6%), nausea (13.3%), difficulty enjoying life (47.2%) and difficulties with self-care activities (6.1%) were less prevalent, but were associated with the greatest likelihood of reporting poorer QoL.	Cancer-Specific Medical Factors	Subtheme: Living with Loss ‘I can’t drive anymore, which I really miss … I feel my whole independence is gone and that’s the saddest thing for me anyway’. [ESM073] ‘I have constant diarrhea … so I never like to be too far from a toilet … and that has affected my confidence in going out, it would make me more inclined to stay in … I am probably giving up on a social life sooner than I should be’. [RSM028]	This subtheme describes the impact of survivorship issues on CRC survivors’ QoL, and predominantly focused on less prevalent issues of bowel dysfunction, peripheral neuropathy, fear of recurrence, financial difficulties, employment difficulties and impact of cancer on the family. The specific impacts of these issues on survivors’ daily lives were linked to factors associated with the greatest likelihood of reporting poorer QoL, including feeling ill, difficulties enjoying life and engaging in self-care.
Satisfaction with continuity of care, unmet information needs, type of hospital attended, having a treatment summary, having access to a named nurse or doctor for advice about cancer-related concerns and feeling supported by a community/primary care healthcare providers were associated with QoL for physical, social, emotional, functional, CRC-specific and overall QoL in univariate and/or multivariate analysis.	Healthcare factors associated with poorer outcomes for one or more QoL domains (p ≤ .05): • Access to a named doctor • Access to a named nurse • Continuity of care • Feels supported by community/primary care • Treatment summary • Type of hospital attended • Unmet information needs 95.1% (n = 272) had access to a doctor, nurse or other healthcare professional to discuss concerns about their cancer or survivorship issues. 95.1% (n = 269) felt supported by hospital staff. 76.2% (n = 215) felt supported by primary care professionals. 61.1% (n = 171) felt supported by community services staff. 67.6% (n = 184) reported one or more unmet information needs.	Healthcare System	Subtheme: Feeling Known and Valued by Oncology Professionals ‘Regardless of what it is, [they] straighten [it] out for me … you can see that they are caring, anything that you have to do, they’re right here, they’re right behind you’. [ESM003]	This subtheme described the characteristics of relationships between CRC survivors and healthcare professionals.
			Subtheme: Ebb and Flow of Support in Primary and Community Care ‘He [The General Practitioner] just sat there, he wasn’t looking at the screen, he was doing nothing, he just swiveled his chair around, and he said, “[Your husband] is worried about you. You don’t want to have him worried about you, tell me what happened”’. [RCM001] ‘One place where I got no support, and it really hurt me a lot, was my local [general practitioner]. I really felt that. I was shocked, so much so that I just haven’t the heart to go back there. So, I’m in limbo now, because I don’t know where to go [about my fatigue]’. [ESM158]	This subtheme describes the nature of supportive care provided to CRC survivors by primary care practitioners and community healthcare workers.
			Subtheme: The Impact of Power Imbalances in the Healthcare Relationship ‘The door of the unit was open, and all I could hear from inside was – ‘We’re so busy today’ … eventually, the clinical nurse manager came out and sat down beside me and proceeded to hold the consultation in the waiting room … I had been thinking myself, are you so diminished by all of this that you’re not able to stand up for yourself? … it was insulting’. [RCM001]	This subtheme describes the adverse effects of paternalistic approaches to healthcare provision on CRC survivors’ QoL, dignity and autonomy in decision-making.
			Subtheme: Managing Survivorship Issues: What Drives Survivors’ Perceptions of Supportive Care? ‘For the neuropathy, I felt I was just being pushed from one to another, I’d come out after being in the clinic, and there was nothing at the end of it … and then that’s it for another six months … that’s very distressing’. [ESM073] ‘I think [sexual dysfunction] should be brought up … It just never came up, I’ve no problem bringing it up if I can remember when I meet them, but then there’s so few times that you meet them’. [ESM006]	This subtheme highlights the issues that influenced cancer survivors’ perceptions of supportive care and experiences of unmet needs, including barriers to supportive care.
			Subtheme: Contrasts Between Preparation For Life During and After Cancer ‘Access to the psychologist was only available for the duration [of treatment] … There wasn’t any follow-up … when [treatment] ended, that was it, “You’re done now”’ [ESM043] ‘They gave me practically no information as to what would happen after the chemotherapy the way I would feel … I remember thinking the last day I walked out … “Gosh, I don’t know anything now, what’s going to happen now?”’ [ESM158]	This subtheme describes the challenges of transitioning from the end of cancer treatment to follow-up care, describing inadequate preparation for long-term effects of cancer treatment.
Accessing voluntary cancer support services was associated with poorer QoL for physical, CRC-specific and overall QoL in univariate and/or multivariate analysis. Participants primarily accessed nurses or doctors to discuss cancer-related worries. Few participants accessed voluntary cancer support services.	Health efficacy factors associated with poorer outcomes for one or more QoL domains (p ≤ .05): • Accessing voluntary cancer support Services 80.8% (n = 227) had access to a named nurse. 79.9% (n = 226) had access to a named doctor. 14.9% (n = 42) accessed voluntary cancer support services. 22.2% (n = 49) accessed online services for support or information. The most common reasons for not engaging with voluntary cancer support services included having adequate support from one’s family (17.2%, n = 33), being unaware of support services that were available (13.5%, n = 26), having adequate support from friends (8.9%, n = 17) or healthcare professionals (8.3%, n = 16).	Health Efficacy	Subtheme: Striving for Vigilance ‘I’m certainly more keen on maintaining a position where perhaps the symptom might come up and you notice very early … and not doubting myself … I’m making a note of something and make sure that I am seeing somebody’. [ESM043] ‘You can get a bit paranoid and then you start thinking you’re bothering people. Ringing up the nurses to get my bloods done … just to check, all that kind of thing’. [ESM043] ‘I’ve spent a fortune on consultants, my doctor is fed up listening to me and it just turns you into someone like that’. [ESM036]	This subtheme highlights cancer survivors’ engagement with cancer surveillance. While vigilance was reassuring, supporting early detection of recurrent disease, it was balanced with fears of compromising follow-up care if they were perceived to be bothering healthcare professionals unnecessarily.
Ethnicity was a controlling factor associated with QoL in the functional well-being domain in multivariate analysis. Between 20% and 48% reported difficulties with relationships and support from their family members and friends which influenced their QoL.	Cultural factors supporting prediction of poorer outcomes for one or more QoL domains (p ≥ .05): • Ethnicity (FWB) 34.3% (n = 94) reported difficulties with emotional support from their family members, which was associated with poorer QoL outcomes (OR = 7.155; CI: 3.971–12.890; p ≤ .0015). 23.0% (n=51) reported difficulties feeling close to their partner, which was associated withpoorer QoL outcomes (OR=6.364; CI: 2.987–13.559; p ≤ .001). 20.0% (n = 56) reported difficulties with family members’ acceptance of their illness, which was associated with poorer QoL outcomes (OR = 26.304; CI = 7.975–86.761; p ≤ .001). 43.0% (n = 120) reported difficulties feeling close to their friends, which was associated with poorer QoL outcomes (OR = 7.044; CI = 4.119–12.048; p ≤ .001).47.8% (n = 129) reported difficulties with emotional support from their friends, which was associated with poorer QoL outcomes (OR = 8.569; CI: 4.948–14.838; p ≤ .001).	Cultural Context	Subtheme: Weaving a Safety Net of Support in Survivorship ‘My wife was fantastic, and even the kids were great … because they kept bringing you back to reality … making life easier for me … I got this huge supportive feeling from [my wife], she was always there; I always had somebody’. [ESM005]	This subtheme described the Safety Net(work)s of Support that CRC survivors accessed beyond the healthcare system to bridge the gaps of needs which were not sufficiently addressed by the healthcare system.
		Cultural Context	Subtheme: Terms and Conditions of The Safety Net of Support: Visibility, Availability, Stigma and Skepticism ‘I have, what I thought were close friends who just never bothered … I think that they must think that you don’t have any issues afterwards that you just come through it’. [ESM036] ‘There was plenty of groups for breast cancer and stuff like that, but not for bowel … With the symptoms I was having, it would be better because you’d have someone else who might have the same system or might have found some way to cope with that’. [RSM018]	This subtheme describes the factors which influenced the impact and supportiveness of the networks of support accessed by cancer survivors to address their unmet needs. The invisibility and stigma associated with late and long-term effects of CRC could hinder the quality of support received from the safety network of support.
Participants reported reasonably positive social and emotional well-being on FACT-C subscales. However, between 30% and 51% reported psychosocial distress, which influenced their QoL.	Mean social well-being scores = 23.3 (SD = 5.4). Mean emotional well-being score = 20.1 (SD = 3.9). 29.9% (n = 88) reported symptoms of anxiety or depression, which was associated with poorer QoL outcomes (OR = 6.066; CI = 3.337–11.028; p ≤ .001). 51.3% (n = 143) were worried about their condition getting worse, which was associated with poorer QoL outcomes (OR = 2.917; CI = 1.793–4.744; p ≤ .001). 43.7% (n = 121) were worried about their condition getting worse, which was associated with poorer QoL outcomes (OR = 3.010; CI = 1.836–4.933; p ≤ .001). 42.0% (n = 116) reported feeling sad, which was associated with poorer QoL outcomes (OR=8.579; CI = 4.926–14.938; p ≤ .001). 43.8% (n = 126) were satisfied with their cancer-related coping, which was associated with poorer QoL outcomes (OR=5.264; CI=3.160–8.770; p ≤ .001). 40.1% (n=110) reported feeling nervous, which was associated with poorer QoL outcomes (OR = 4.455; CI=2.645–7.503; p ≤ .001).	Psychological Well-being	Subtheme: Living in the Shadow of CRC ‘I’ve gone through this thing, but like, do you ever go through it? No, you’re still going through it … It’s always there, so the cancer thing is always there’. [ESM043] ‘I’d probably end up in a wheelchair if I got more chemo. So that’s my biggest worry that it would come back, and I couldn’t have treatment’. [ESM073] ‘While I had no symptoms of bowel cancer before the surgery, I have had all the symptoms that you are told to look out for since the surgery. That is, I have constant diarrhea, I would have maybe five, six, maybe even seven bowel movements a day’. [RSM028] ‘You’ll never have a headache again, it’ll be a brain tumor … you’ll think worst case scenario, and that is me, I have turned into that person … you wake in the morning, how am I? … Am I okay? … Is that different from one I’ve had before? … Which of the symptoms is that now? Which thing is that now?’ [ESM036]	This subtheme encompasses the psychological impact of cancer diagnosis and treatment, particularly fear of recurrence.
		Psychological Well-being	Subtheme: Striving to Find Benefits in the Experience of Cancer ‘I don’t know how it affects other people, there was a woman down the road there, she’s dead now, God rest her, but she got neuropathy after cancer, she could hardly walk, she got it really bad. I’m counting my blessings; it’s not stopping me getting around’. [PCM015] ‘I do some unofficial home visits to neighbors and friends who are diagnosed with CRC and use a stoma. I tell them what works for me and keep the chat very upbeat and positive. I tell them they can ask me anything and they do because we can be empathetic’. [RCM001] ‘I did put the word out in the oncology ward, I told them in there … that if anybody wanted to talk about it or was due to go through it and wanted to talk about it on a one-to-one basis that I would be quite willing to do that’. [RCM013] ‘I have told all of my friends, a lot of whom were invited [to BowelScreen] but hadn’t bothered filling in the form … so I’ve been advocating to every one of the age to make sure that they do that’. [PCM018]	This subtheme describes the strategies utilized by CRC survivors to cope with and reframe the impact of cancer and its treatment.

Demographic Characteristics

The quantitative and qualitative findings converged regarding the impact of age on QoL. Younger CRC survivors consistently reported poorer emotional, social and CRC-specific well-being. This finding was supported and explained by the findings of the qualitative phase, where younger CRC survivors described age-related disparities in care, as services were designed to address the needs of an older age group who were more likely to be diagnosed with CRC. As a result, younger survivors felt their cancer-related needs related to their life stage were not addressed, and limited supports were available beyond the health system.

Socio-Ecological Factors

Socio-ecological factors are comprised of socio-economic status, social support and life burden, where daily stressors compromise a persons’ ability to cope with unanticipated circumstances. Quantitative and qualitative results converged to quantify and explain the financial impact of cancer which persisted beyond the period of treatment. Financial difficulties were associated with poorer QoL and were reported by one-quarter (25.7%) of questionnaire participants. Interviews highlighted difficulties survivors experienced in returning to work and shortcomings in financial support available from social services.

General Health Factors

Within the general health domain, health and comorbidity are important considerations in modelling cancer survivors’ QoL outcomes. Converging quantitative and qualitative findings indicate that comorbidity and disease status were associated with poorer QoL. However, qualitative data described the confounding effect of ageing and comorbidity upon QoL. In particular, cancer survivors attributed survivorship issues such as fatigue, changes in cognitive function and sexual dysfunction to the process of ageing or comorbid conditions.

Cancer-Specific Medical Factors

Aligning with the Ashing-Giwa model, diagnosis, disease status and previous treatments were each associated with cancer survivors’ QoL in multivariate analysis. Regarding the long-term effects of cancer and its treatment, quantitative and qualitative data converged, describing the direct link between survivorship issues and cancer survivors’ QoL outcomes. While physical, psychological and social survivorship issues were each individually associated with poorer QoL, quantitative analysis suggested functional decline and social upheavals reported by a small number of participants were associated with the highest likelihood of reporting poorer QoL. These difficulties included feeling ill (12.3%), difficulties with self-care activities (6.1%), dissatisfaction with QoL (54%), difficulty enjoying life (47.2%) and family members’ difficulty accepting their loved ones’ diagnosis (20%).

During interviews, CRC survivors’ narratives provided an enhanced understanding of how less prevalent survivorship issues mediated the relationship between physical symptoms attributed to cancer treatments and QoL. Participants’ narratives focused on the impact of survivorship issues, with issues related to peripheral neuropathy, bowel dysfunction, fear of recurrence, financial toxicity, employment difficulties and impact of cancer on the family identified as common sources of functional difficulty and psychosocial distress. Survivors described how survivorship issues impacted on multiple domains of their QoL and contributed to their feeling unwell, feeling dissatisfied with their QoL and having difficulties with independence and enjoying life, which they attributed to greater psychological distress and social isolation.

Healthcare System

Healthcare system factors include access to healthcare, quality of healthcare and quality of relationships with healthcare professionals. The quantitative and qualitative data converged to confirm, explain and expand understanding of the impact of healthcare systems on CRC survivors’ QoL outcomes. Continuity of care was a consistent predictor of CRC survivors’ QoL. Although health insurance was associated with differences in QoL outcomes, cancer survivors did not describe any differences in perceptions of access to or support from healthcare professionals in the post-treatment period based on their health insurance status or location of care (public vs. private healthcare systems) during qualitative interviews.

Healthcare professionals were described as accessible, and 95.1% of participants (n = 272) had access to a doctor, nurse or other healthcare professional to discuss concerns about their cancer or survivorship issues. However, characteristics of the relationship between participants and healthcare professionals in various settings could influence QoL, including the quality of relationships, feeling known and supported by healthcare professionals and feeling informed about the long-term implications of cancer and its treatment. Feeling supported by primary and community care professionals was associated with physical well-being in multivariate analyses, but fewer participants reported support from these sources (61.1%–76.2%) compared to hospital-based professionals (95.1%). Qualitative data suggested that access to support from primary and community care professionals was inconsistent. For participants who received support from primary or community care professionals, psychosocial support from these professionals was of greatest significance and value to CRC survivors. While quantitative data suggested that specialist oncology professionals were accessible to cancer survivors, several survivors described difficulties accessing supportive care to address physical effects of cancer treatment. Perceptions of support from specialist oncology professionals appeared to be influenced by the location of care, resources available and whether pathways to access support were direct to the professional or indirectly through another service or referral.

Health Efficacy

Within the Ashing-Giwa model, health efficacy manifests as survivors’ health practices and engagement with healthcare services. Quantitative and qualitative data converged to quantify and explain patterns of health and support service utilization within the sample. The quantitative phase identified multiple sources of information and support to address cancer-related concerns; the majority of participants had access to a nurse (80.8%) or doctor (79.9%), who represented the primary sources of information and support. Few participants accessed voluntary cancer support (14.9%) or online information or support (22.2%). Having access to a named doctor was associated with more positive CRC-related well-being and overall QoL in multivariate analysis, while having access to a named nurse was associated with more positive emotional well-being. Notably, feeling supported by hospital staff was not associated with QoL outcomes in multivariate analysis. Meanwhile those who accessed voluntary cancer support services were more likely to report poorer physical, CRC-specific and overall QoL.

Qualitative data provided an enhanced understanding of the factors driving survivors’ engagement with healthcare, voluntary and online support services. Survivors expressed a desire to be both compliant with the recommendations of healthcare professionals and also autonomous, exercising their knowledge of their condition to self-manage survivorship issues and make decisions about their care. Many participants endeavoured to self-manage survivorship issues and self-navigate the healthcare system and avoid contacting healthcare professionals unless absolutely necessary. Qualitative data suggested that survivors’ experiences of late effects contributed to unmet information and supportive care needs, which were mediated by survivors’ perceptions of accessibility, supportiveness and continuity of healthcare services, which in turn affected QoL.

Cultural Context

The cultural context of the Ashing-Giwa model focuses on ethnicity, language, social networks and quality of family and social relationships. Qualitative data provided greater insight to the mechanisms by which healthcare system factors and health efficacy factors could influence QoL. Family members, friends and communities were central to participants’ well-being and feeling supported. Cancer survivors who described experiencing unmet needs and shortcomings in healthcare services attempted to address these needs, describing networks of support they developed beyond the healthcare system. These networks of support encompassed family members, social networks and cancer peers and were conceptualized as a safety net(work), which bridged the gap of unmet needs.

Narratives regarding support from family members were overwhelmingly positive. However, support from wider social networks, voluntary services and online supports were subject to terms and conditions and cultural barriers. Integration of quantitative and qualitative data identified some divergence in experiences of accessing and using the safety net(work) of support. While qualitative data suggested familial support was unconditional, quantitative data in contrast, revealed that in reality, many survivors experienced difficulties with the quality of relationships and nature of support from their family members (20.0% and 34.3%, respectively). More than two-fifths of questionnaire participants described difficulties with the quality of relationship and nature of support received from friends (43.0% and 47.8%, respectively). Qualitative data highlighted barriers to support from social networks, including the invisibility of CRC, as its treatment did not manifest outward indicators of a cancer diagnosis such as alopecia or significant weight loss.

Voluntary and peer cancer support services were not widely accessed by questionnaire participants (14.9%). Quantitative data suggested that accessing voluntary and online support services was affected by the survivors’ cultural context, including the accessibility and local availability of services, levels of support from family, friends and wider community. Qualitative and quantitative data converged, confirming and further explaining how cultural factors could further influence survivors’ utilization of voluntary, peer and online services. Stigma associated with discussion of bowel dysfunction, visibility of survivorship issues, availability of supports in local communities and scepticism of online resources limited the range of supports available to cancer survivors and their openness to seeking support from social networks and generic cancer support services.

Psychological Well-being

Psychological well-being is concerned with psychological functioning and the experiences of psychological distress, depression and anxiety. While quantitative data suggested cancer survivors experienced reasonably positive psychological and social well-being, between 30% and 51% of survivors described psychological issues related to fear of recurrence, anxiety and depression. Convergent qualitative data highlighted the complexity of cancer survivors’ fears and emotional distress and the factors driving their fears. Fear of recurrence was described as one of the most difficult to manage survivorship issues, driven by fears of reliving the experience of cancer, of difficulties differentiating between symptoms of recurrence and non-threatening symptoms, and the potential for living with worsening effects of cancer treatments. However, qualitative data contained rich descriptions of the strategies cancer survivors used to cope with and derive benefit from the experience of cancer. Several survivors compared themselves to others with more complex illnesses; others described altruistic motivations for providing support to others affected by cancer. In deriving benefits from the experience, survivors consistently described the positive impact of cancer on the closeness of relationships within families.

The Model of Healthcare Factors Influencing Quality of Life in Cancer Survivorship

The meta-inferences yielded from PIP analysis underpin the M odel o f H ealthcare f a ctors influencing Q uality of Life in C ancer S urvivorship (MoHaQ-CS) (Figure 5). Key findings from this research suggest that late effects of CRC are integral to survivors’ QoL, conceptualized as a price to pay for survival. Self-management of late effects mediated survivors’ utilization of healthcare services; where survivors were able to self-manage late effects, and overcome effects which compromised their functional well-being, QoL was described positively. Where survivors encountered difficulties self-managing late effects, the experience contributed to unmet information and supportive care needs, which mediated survivors’ perceptions of access to, support from and continuity of healthcare, in turn affecting QoL. Difficulties self-managing late effects prompted survivors to seek additional support from healthcare professionals. Where survivors experienced continuity in care, appropriate support to self-manage late effects, and experienced relief or mastery of symptoms, QoL was positively influenced.

Figure 5.

The Model of Healthcare factors influencing Quality of Life in Cancer Survivorship (MoHaQ-CS).

For some survivors, their encounters with the healthcare system did not meet their expectations for support, and these encounters contributed to psychological distress, arising from interpersonal interactions or perceived shortcomings in support. Where the support received from healthcare services was inadequate, this manifested and exacerbated unmet information and supportive care needs. In these circumstances, survivors sought further support beyond the realms of formal oncology healthcare services to address unmet needs. Survivors attempted to Bridge the Gap of Unmet Need through a Network of Supports Beyond the Healthcare System, encompassing their family members, their social network, cancer advocacy services and online resources. However, the nature of the Network of Supports Beyond the Healthcare System were dependent upon the cultural context in which the individual resided and received healthcare, as beliefs relating to survivors’ acculturation, interconnectedness and worldview could hinder access to psychosocial support. Survivors’ ability to cultivate a support network directly and indirectly influenced QoL. Positive experiences of support from support networks could positively affect QoL and provide practical support or information to self-manage late effects. However, even where survivors identified sources of support, the accessibility and effectiveness of this support could be influenced by the quality of relationships, worries about burdening or distressing others, social stigmas about the discussion of embarrassing late effects, difficulties discerning the trustworthiness of online resources, lack of local cancer support services specific to CRC or a belief that cancer support was not required.

Discussion

The MoHaQ-CS represents the processes by which healthcare factors may influence CRC survivors’ QoL. The joint display (Table 5) illustrates the integration of quantitative and qualitative findings to address the study aim, in line with recommended approaches to integration (Creswell and Clark, 2017; Guetterman et al., 2015). The adapted model incorporates the quantitative factors which predicted CRC survivors’ QoL within the domains of the Ashing-Giwa Contextual Model of Health-Related Quality of Life (Figure 3). The qualitative findings confirm, explain and extend understanding of the statistical factors influencing QoL, identifying the pathways by which healthcare experiences may directly and indirectly affect cancer survivors’ QoL (Figure 4).

Contribution to the Field of Mixed Methods

This paper adds to the methodological discourse of MMR, illustrating the use of an explanatory sequential MMR design to build and test a theoretical framework, the MoHaQ-CS, responding to a reported gap in the literature (Evans et al., 2011; Kivunja, 2018). Using an established conceptual model to guide exploratory research can support the development of new and more nuanced understanding of complex phenomena; this knowledge may subsequently support the development of interventions and services which can more effectively respond to the needs of people living with chronic diseases (Lynch et al., 2018). Moreover, use of an a priori conceptual model, integrated throughout the study from conception, through design, analysis and interpretation has been found to support integration in MMR (Evans et al., 2011). The Ashing-Giwa (2005) Contextual Model of Health-Related Quality of Life was selected to guide this study as it could support an exploration of the complex factors which may influence QoL in cancer survivorship, including those derived from the healthcare context of particular interest to this study. The Ashing-Giwa (2005) model informed key decisions in the design of this study, including the selection of methods, the sampling strategy and data collection methods and provided a framework for sequential and subsequent integrated analysis of quantitative and qualitative data.

The PIP is a systematic and replicable technique to support integration and visualization of MMR data (Johnson et al., 2019). The implementation of PIP in this study has been modified, adopting a deductive analytical approach which leverages the use of a theoretical framework to guide The Cost of Survival study. PIP has enabled a more nuanced understanding of the variables driving QoL outcomes in cancer survivorship. Integration of qualitative and quantitative findings in this study supported expansion of the Ashing-Giwa (2005) model, for the first time, explaining the pathways by which healthcare systems may affect cancer survivors’ QoL and identifying factors which may potentially confound analysis of QoL outcomes among this population.

Integrating quantitative and qualitative data deductively following PIP illuminated contextual and cultural factors directly and indirectly influencing QoL outcomes. The process of centring the domains of the Ashing-Giwa (2005) model as pillar concepts ensured that analysis remained focused on the relationship between the concepts of interest, healthcare systems and QoL, enabling comparison and contrast between quantitative and qualitative data to illuminate the nuances of the interaction between these concepts. While the product of this study, the MoHaQ-CS, maintains the core domains of the Ashing-Giwa (2005) model, results of this study suggest that self-management skills and cultural context, specifically, the availability, accessibility and effectiveness of support, are critical drivers of healthcare utilization, unmet needs and consequently, QoL in cancer survivorship.

Implications for Cancer Survivorship Care

Given the vision for developing models for cancer survivorship care which integrate virtual or remote follow-up, and shared nurse-led and primary care-led models of care, there is a need for greater understanding of the mechanisms by which healthcare services may impact upon cancer survivors’ QoL outcomes across varied healthcare systems and models of care. The MoHaQ-CS may support the development and adaptation of supportive care interventions which are sensitive to context and the personal needs of people living with and beyond cancer (Richards & Hallberg, 2015). Furthermore, the MoHaQ-CS could inform planning for comprehensive evaluation of factors influencing cancer survivors’ QoL within supportive care interventions and inform future studies to identify risk factors for poorer outcomes and disparities in QoL outcomes.

Limitations

PIP has been proposed as a rigorous, inductive approach supporting systematic integrated analysis and display of MMR data (Johnson et al., 2019). However, to our knowledge, this is the first example of a deductive approach to PIP, and as such, the results of the study should be interpreted in the context of this potential limitation. An inductive approach to integrated analysis may have yielded alternative findings. Given the context in which this study was conducted, the generalizability of the MoHaQ-CS may be limited by the funding and organization of cancer services in Ireland. Firstly, healthcare provision in Ireland operates in a mixture of both public and private settings, which is quite distinct from the privatized and universal models of healthcare operated in other international jurisdictions. Secondly, the structure of cancer-related follow-up care in Ireland is not currently standardized. In private healthcare systems, follow-up healthcare is primarily lead by physicians, while in public healthcare systems, follow-up care is largely nurse-led; however, additional physician-led care may be provided to some cancer survivors.

Conclusion

Integration of MMR data within this study serve to advance the utility of the Ashing-Giwa (2005) Contextual Model of Health-Related Quality of Life with broader populations of cancer survivors in international contexts, cohesively describing the processes by which healthcare factors may influence cancer survivors’ QoL. The use of a theoretical framework to guide exploratory MMR can enable more nuanced understanding of complex phenomena, while supporting high levels of integration in the design, analysis and interpretation of MMR. The use of a modified, deductive PIP to support integration within this study has enabled a focused analysis, allowing comparison and contrast between quantitative and qualitative data to illuminate the nuances of the interaction between the concepts of interest, healthcare system and QoL. The product of this analysis, the MoHaQ-CS can provide the pillars to enable comprehensive development and evaluation of complex interventions which are responsive to the supportive care needs and health-related QoL of cancer survivors.

Footnotes

Acknowledgements

The authors would like to thank the physicians, nurses and staff of the surgery and medical oncology departments of participating hospitals and voluntary cancer support services who facilitated data collection.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

This research was supported by funding from the Health Research Board of Ireland (HPF.2014.715).

Data Availability

The datasets generated during and/or analysed during the current study are not publicly available. Excerpts of data are available from the corresponding author upon reasonable request.

ORCID iD

Amanda Drury

References

Ashing-Giwa

K. T.

(2005). The contextual model of HRQoL: A paradigm for expanding the HRQoL framework. Quality of Life Research, 14(2), 297–307. https://doi.org/10.1007/s11136-004-0729-7

Ashing-Giwa

K. T.

Lim

J. W.

(2008). Predicting health-related quality of life: Testing the contextual model using structural equation modeling. Applied Research in Quality of Life, 3(3), 215–230. https://doi.org/10.1007/s11482-009-9057-y

Ashing-Giwa

K. T.

Lim

J. W.

(2011). Health-related quality of life outcomes among cervical cancer survivors: Examining ethnic and linguistic differences. Cancer Epidemiology, 35(2), 194–201. https://doi.org/10.1016/j.canep.2010.06.006

Bazeley

(2018). Integrating analyses in mixed methods research. Sage Publications Ltd.

Braun

Clarke

(2006). Using thematic analysis in psychology. Qualitative research in psychology, 3(2), 77–101. https://doi.org/10.1191/1478088706qp063oa

Cheng

K. K. F.

Wong

W. H.

Koh

(2016). Unmet needs mediate the relationship between symptoms and quality of life in breast cancer survivors. Supportive Care in Cancer, 24(5), 2025–2033. https://doi.org/10.1007/s00520-015-2994-0

Creswell

J. W.

Clark

V. L. P.

(2017). Designing and conducting mixed methods research. Sage Publications Ltd.

Curry

(2015). Mixed methods in health sciences research: A practical primer. Sage Publications Ltd.

Cuthbert

C. A.

Farragher

J. F.

Hemmelgarn

B. R.

Ding

McKinnon

G. P.

Cheung

W. Y.

(2019). Self-management interventions for cancer survivors: A systematic review and evaluation of intervention content and theories. Psycho-Oncology, 28(11), 2119–2140. https://doi.org/10.1002/pon.5215

10.

Di Fabio

Koller

Nascimbeni

Talarico

Salerni

(2008). Long-term outcome after colorectal cancer resection. Patients' self-reported quality of life, sexual dysfunction and surgeons' awareness of patients' needs. Tumori Journal, 94(1), 30–35. https://doi.org/10.1177/030089160809400107

11.

Department of Health United Kingdom (2010). National Cancer Patient Experiences Survey Programme: 2010 National Survey Report. Quality Health, United Kingdom. https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/216682/dh_122520.pdf

12.

Drury

Payne

Brady

A.-M.

(2017). Cancer Survivorship: Advancing the concept in the context of colorectal cancer. European Journal of Oncology Nursing, 29, 135–147. https://doi.org/10.1016/j.ejon.2017.06.006

13.

Drury

Payne

Brady

A.-M.

(2020a). Colorectal cancer survivors’ quality of life: A qualitative study of unmet need. BMJ Supportive & Palliative Care. https://doi.org/10.1136/bmjspcare-2020-002190

14.

Drury

Payne

Brady

A.-M.

(2020b). Identifying associations between quality of life outcomes and healthcare-related variables among colorectal cancer survivors: A cross-sectional survey study. International Journal of Nursing Studies, 101, 103434. https://doi.org/10.1016/j.ijnurstu.2019.103434

15.

Drury

Payne

Brady

A. M.

(2021). Prevalence vs impact: A mixed methods study of survivorship issues in colorectal cancer. Quality of Life Research. 31(4), 1117–1134. https://doi.org/10.1007/s11136-021-02975-2

16.

Evans

B. C.

Coon

D. W.

Ume

(2011). Use of theoretical frameworks as a pragmatic guide for mixed methods studies: A methodological necessity? Journal of Mixed Methods Research, 5(4), 276–292. https://doi.org/10.1177/1558689811412972

17.

Gordon

L. G.

Patrao

Hawkes

A. L.

(2012). Can colorectal cancer survivors recall their medications and doctor visits reliably? BMC Health Services Research, 12(1), 440. https://doi.org/10.1186/1472-6963-12-440

18.

Guetterman

T. C.

Fetters

M. D.

Creswell

J. W.

(2015). Integrating quantitative and qualitative results in health science mixed methods research through joint displays. The Annals of Family Medicine, 13(6), 554–561. https://doi.org/10.1370/afm.1865

19.

Hadjistavropoulos

Biem

Sharpe

Bourgault-Fagnou

Janzen

(2008). Patient perceptions of hospital discharge: reliability and validity of a Patient Continuity of Care Questionnaire. International Journal for Quality in Health Care, 20(5), 314–323. https://doi.org/10.1093/intqhc/mzn030

20.

Hallett

Pearson

Stephenson

(2014). Adult cancer survivors' experiences of health care interactions and unmet needs in health care services: A systematic review protocol. JBI Database of Systematic Reviews and Implementation Reports, 12(9), 36–44. https://doi.org/10.11124/jbisrir-2014-1459

21.

Hammer

M. J.

Cartwright-Alcarese

Budin

W. C.

(2019). Theoretical frameworks and philosophies of care. In Payne

Murphy-Ende

(Eds.), Current trends in oncology nursing (2nd ed.). Oncology Nursing Society.

22.

Institute of Medicine (2006). From cancer patient to cancer survivor: Lost in transition. The National Academies Press.

23.

Johnson

R. E.

Grove

A. L.

Clarke

(2019). Pillar integration process: A joint display technique to integrate data in mixed methods research. Journal of Mixed Methods Research, 13(3), 301–320. https://doi.org/10.1177/1558689817743108

24.

Kivunja

(2018). Distinguishing between theory, theoretical framework, and conceptual framework: A systematic review of lessons from the field. International Journal of Higher Education, 7(6), 44–53. https://doi.org/10.5430/ijhe.v7n6p44

25.

Lynch

E. A.

Mudge

Knowles

Kitson

A. L.

Hunter

S. C.

Harvey

(2018). There is nothing so practical as a good theory”: A pragmatic guide for selecting theoretical approaches for implementation projects. BMC Health Services Research, 18(1), 857. https://doi.org/10.1186/s12913-018-3671-z

26.

O'Cathain

Murphy

Nicholl

(2007). Why, and how, mixed methods research is undertaken in health services research in england: A mixed methods study. BMC Health Services Research, 7(1), 85–96. https://doi.org/10.1186/1472-6963-7-85

27.

Quality Health (2012). Quality of Life of Cancer Survivors in England: Report on a Pilot Survey using Patient Reported Outcome Measures (PROMS).Department of Health, United Kingdom. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/267042/9284-TSO-2900701-PROMS-1.pdf

28.

Richards

D. A.

Hallberg

I. R.

(2015). Complex interventions in health: An overview of research methods. Routledge.

29.

Sisler

J. J.

Taylor-Brown

Nugent

Bell

Khawaja

Czaykowski

Wirtzfeld

Park

Ahmed

(2012). Continuity of care of colorectal cancer survivors at the end of treatment: The oncology-primary care interface. Journal of Cancer Survivorship, 6(4), 468–475. https://doi.org/10.1007/s11764-012-0235-9

30.

Snyder

C. F.

Earle

C. C.

Herbert

R. J.

Neville

B. A.

Blackford

A. L.

Frick

K. D.

(2008). Trends in follow-up and preventive care for colorectal cancer survivors. Journal of General Internal Medicine, 23(3), 254–259. https://doi.org/10.1007/s11606-007-0497-5

31.

Tofthagen

(2010). Surviving chemotherapy for colon cancer and living with the consequences. Journal of Palliative Medicine, 13(11), 1389–1391. https://doi.org/10.1089/jpm.2010.0124

32.

Ward

W. L.

Hahn

E. A.

Hernandez

Tulsky

D. S.

Cella

(1999). Reliability and validity of the functional assessment of cancer therapy-colorectal (FACT-C) quality of life instrument. Quality of Life Research, 8(3), 181–195. https://doi.org/10.1023/a:1008821826499

33.

World Health Organisation Quality of Life Group (1995). The world health organization quality of life assessment (WHOQOL): Position paper from the world health organization. Social Science & Medicine, 41(10), 1403–1409. https://doi.org/10.1016/0277-9536(95)00112-K

34.

Wright

Smith

A. B.

Keding

Velikova

(2011). The social difficulties inventory (SDI): Development of subscales and scoring guidance for staff. Psycho-Oncology, 20(1), 36–43. https://doi.org/10.1002/pon.1705

Adapting the Pillar Integration Process for Theory Development: The Theoretical Model of Healthcare Factors Influencing Quality of Life in Cancer Survivorship

Abstract

Keywords

Introduction

Study Background

Methods

Data Analysis and Integration

Results

Sample Characteristics

Integrated Findings

Demographic Characteristics

Socio-Ecological Factors

General Health Factors

Cancer-Specific Medical Factors

Healthcare System

Health Efficacy

Cultural Context

Psychological Well-being

The Model of Healthcare Factors Influencing Quality of Life in Cancer Survivorship

Discussion

Contribution to the Field of Mixed Methods

Implications for Cancer Survivorship Care

Limitations

Conclusion

Footnotes

Acknowledgements

Declaration of Conflicting Interests

Funding

Data Availability

ORCID iD

References