SEARCH

SEARCH BY CITATION

Keywords:

  • adaptive;
  • longitudinal;
  • mixed methods;
  • qualitative;
  • strategies;
  • stroke

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Objectives of this paper
  5. Methods
  6. Summary
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

Background

Qualitative data can add value and understanding to more traditional epidemiological studies. This study was designed to complement the quantitative data from the incidence study the Auckland Regional Community Stroke Study or ARCOS-IV by using qualitative methods to uncover the richer detail of life as a stroke survivor, thereby extending our understanding of the impact of stroke.

Aims

The aims of the study were to identify how the experience of recovery and adaptation changes over time after stroke; and to elicit the strategies people with stroke and their whānau/family use and find helpful in living life after stroke. The aim of this paper is to describe the methodology and also the challenges and advantages of embedding qualitative research into a large epidemiological study.

Methods

Longitudinal study utilizing a Qualitative Description design in a subset of those taking part in the incidence study. Participants will be interviewed at 6, 12, 24, and 36 months after stroke. Semistructured interviews will explore three key areas: (1) issues of importance to people following a stroke and their whānau/family; (2) the perceived impact on people's sense of recovery, adaptation, and hopes; and (3) key strategies that people with stroke and their whānau/family use and find most helpful in living life after stroke. Thematic analysis will be conducted using iterative constant comparative methods.

Conclusions

This methodology paper demonstrates the application of mixed methods in epidemiology. It also considers some of the practical and methodological issues that have emerged and may provide a useful framework for other qualitative projects in population-based studies.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Objectives of this paper
  5. Methods
  6. Summary
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

Despite the valuable data provided by standardized quantitative assessments, many experiences of health and ill-health remain untapped by current measures [1-3]. Though major advances have been made in the field of patient-reported outcomes such as quality of life [4-6], aspects of importance to stroke survivors regarding their recovery and adaptation remain unexplored [7-9]. An understanding of what it is to ‘live a life after stroke’ is important [10-12], not just because highlighting this perspective is morally justifiable but because such views can inform future health-care innovation [13, 14].

A further issue with stroke research is that it has focused largely on the problems of stroke, with the barriers to recovery well described [15-17]. To date, there has been little attention to positive strategies individuals and their whānau/family use to cope with life after stroke [15, 18] or positive life changes following stroke [19]. ‘Strengths-based’ approaches are increasingly identified as a theoretically sound approach to working with people within the mental health sector but as yet have not been well explored in rehabilitation [20, 21]. This study aims to identify the active and adaptive ways in which people respond to stroke utilizing such a strengths-based framework [20, 22, 23] to generate and refine interview questions and to aid data analysis. In addition, the current study is the first we are aware of that will examine recovery and adaptation over the first three-years after stroke.

This paper describes the qualitative component of a large epidemiological study of stroke incidence and outcomes the Auckland Regional Community Stroke Study (ARCOS IV) [24]. The study addresses three fundamental areas where knowledge is currently limited:

  1. issues of importance to people with stroke and their whānau/family that are less amenable to ‘measurement’ including the perceived impact of stroke, and of stroke services, on people's sense of recovery, adaptation, and hopes for the future;
  2. how the experience of recovery and adaptation changes over time and in relation to key variables of interest; and
  3. key strategies that people with stroke and their whānau/family find helpful in living life after stroke.
  4. Qualitative findings will augment the analyses of the full study and be used to inform health care and support interventions for stroke survivors/their families.

Objectives of this paper

  1. Top of page
  2. Abstract
  3. Introduction
  4. Objectives of this paper
  5. Methods
  6. Summary
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

Though it is increasingly common to publish protocols for randomized controlled trials, it is not typical to do this in qualitative research as methods evolve in response to the data with sampling, analysis, and interview questions developed concurrently. We considered it would be most valuable and appropriate to focus our report on the challenges to date of conducting a large longitudinal qualitative study that is informing, and being informed by, an epidemiological study as well as detail our response to those challenges. We are currently 12 months into the study. The six-months interviews have been completed, and the 12 months interviews are currently underway. Thus in this paper we aim to

  1. give background to the study, describing our methodology and the contributions that qualitative research can make to epidemiological studies;
  2. outline the factors and issues that have arisen in implementing our design, how we have addressed them and what we have learned so far; and
  3. illustrate how quantitative and qualitative components of a linked study can bring value not only during analysis of results but also while the research is being conducted.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Objectives of this paper
  5. Methods
  6. Summary
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

This study used a qualitative description methodology [25] informed by Grounded Theory [26, 27]. A specific application for ethical approval for this study was granted by the Northern X Regional Ethics Committee (NTX/11/07/058).

Recruitment

Participants recruited into the ARCOS IV incidence study [24] were asked if they would be interested in receiving information about this study at 28 days after stroke where possible, or 6 months following their stroke where ARCOS researchers considered it burdensome or were not able to ask earlier. Those who were interested in taking part in this study were asked for their consent to share the demographic and medical information collected as part of the incidence study with the qualitative study team to reduce repetition and burden.

Participant eligibility was based on the following criteria:

Inclusion criteria

(1) People ≥16 years of age who experience a stroke as defined according to the World Health Organization criteria, between 1 March 2011 and 29 February 2012; (2) a resident of the Auckland region; and (3) a registered participant of the parent incidence study. All participants who agreed to hear more about this study were contacted by the qualitative team.

Exclusion criteria

(1) Unable to provide informed consent; (2) unable to engage with an interviewer despite support (e.g. even with support from an interpreter or speech and language therapist); and (3) participating in the Motivational Interviewing in Stroke trial or MIST [28].

Sampling

This study aimed to recruit a sample of 40 participants with stroke and a nominated whānau/family member where possible. Whānau is a Māori language word commonly used in New Zealand English and is used here to mean extended family. In practical terms, that has meant an adult of importance to the participant, who is supporting them in their recovery. Given the nature of people with stroke, and prior research, it was anticipated that n = 40 would enable data saturation [29] (i.e. no new data emerging) in relation to key questions and allow for 25% attrition over time. Moderate sample sizes of 20–50 are usual for Qualitative Description studies [30], where maximum variation of experience is sought rather than recruitment of a representative sample. Accordingly, people with people with communication and cognitive impairments are eligible to participate in the study to ensure a range of perspectives and experiences are captured. Their interviews are shorter and participants often chose to be interviewed with their nominated family member for support with word finding and recall of events. Questions are more directed than open ended to reduce anxiety.

Data collection

Each participant and their nominated whānau/family member (if identified and consenting) are being interviewed at 6, 12, 24, and 36 months using minimally structured open-ended questions [25]. The interview guide is attached (Appendix S1). Interviews have been conducted individually or as a dyad, according to the participants' preference and audiotaped and transcribed verbatim. Emotional expression such as laughter, tearfulness, and emphasis are indicated in the transcript [31]. Field notes are also completed during and after the interviews [32].

The research officer carrying out the interviews (S. R.) is an occupational therapist and has previously worked in stroke services. While this experience provides background, context, and some shared experience with participants, it could also influence interview questions, interactions, and analysis. This is being managed through written reflections and journaling, supervision, and a team approach to coding.

Analysis plan

Overview of the approach

The Qualitative Description approach aims to produce a nuanced summary and description of an experience. While it is less common to articulate the theoretical basis for the methods in epidemiological research, such transparency is crucial in qualitative research as it allows others to understand the frame of reference, or context for interpretation. The longitudinal design of this Qualitative Description Study [25] assumes that life after stroke is socially constructed or, that the experience results from an interaction of many factors involving the person, and their social context and relationships [27]. Thus, social constructionism informs our design, with the data explored in relation to predefined areas identified in the study aims. Because Qualitative Description research does not have a specific methodology, researchers often make use of methods or techniques from other traditions [25]. To ensure a rigorous approach, our methods draw on the principles of grounded theory [33, 34]. Grounded theory methods can be used flexibly to inform the ongoing direction of a study and provide additional tools for data analysis. For example, comparisons during each stage of analysis including data with data within interviews, early interviews with later ones, and between participants [35].

Thematic analysis is being conducted concurrently with ongoing interviews, using iterative constant comparison methods [36]. Hence, data from the first five participants were analyzed prior to completing the subsequent interviews and this process will be ongoing throughout the study. As such, the aim of analysis is to identify key themes that characterize people's experiences of recovery and adaptation and key issues for people in relation to important life roles and their strategies. This is achieved by staying close to the data and language of the participants. It involves some interpretation but does not generate theory or concepts [37]. The intention is not simply to generate a list of strategies or describe coping approaches but shed light on the process by which people reconstruct a life after stroke.

Coding

Each transcript is being analyzed individually. Contextual and demographic information, whether the data comes from the participant or whānau/family member, is indicated in the coding. The core team for the study (S. R., A. T., and K. M.) meet regularly to discuss analysis and areas for further exploration in subsequent interviews. Codes are being developed from the data itself, with the process influenced by new data and emerging insights [25].

Data will also be explored for changes over time and for both similarities and differences according to age, gender, ethnicity, and other factors emerging in the data. Though there are commercial analysis packages available [38-40] we have developed a study-specific database to facilitate storage and coding of data. This has been done to assist analyses with such a large data set, enable the generation of reports both within and across participants, and to facilitate observations over time.

Methodological and ethical considerations and actions

The key issues that have arisen to date are specified below, and the actions we have taken to deal with them are delineated and summarized in Table 1.

Table 1. Summary of challenges and strategies for resolution
ChallengesStrategies for resolution
Researcher perspectives of participant burden impacting on recruitmentEducation and discussion of ethical and methodological consequences in approaching research participants and enabling them to make a choice [41, 42]. Regular communication between researchers on both teams. Sharing of emerging findings and potential impact on practice. Sensitive issues training [50, 52-54].
Reducing participant burden and maximizing benefitIncrease comfort by efforts to create rapport, trust, empathy, and understanding. Responsible questioning [58], monitoring for distress, onward referral when appropriate, attention to sound methodology, and thoughtful dissemination [59]. Use of a community reference group to inform translation of findings.
Ensuring diversity of experienceRegular review of sample demographics. Concurrent recruitment and analysis [25]. Ethical amendment to widen sources of recruitment where needed.
Facilitating participant retentionRegular study updates, recording of comprehensive contact information [63], and emphasis of the importance of the study and participant contribution.
Managing large amounts of dataUse of a specifically designed data base. Sharing of strategies across similar studies in other populations being carried out by other researchers.
Researcher perspectives of participant burden

While vulnerable people need to be protected from unnecessary and distressing effects of research, benevolence needs to be balanced with paternalistic power [41, 42]. Gate keeping, inaccurate interpretation of eligibility, judgments about participants' ability to understand the research, and assumptions regarding the burden on participants can have a significant effect on recruitment into research [43, 44]. However, if the frail, complex, and difficult to engage are always excluded from research, guidelines for best practice can never be developed with any certainty that they are appropriate and acceptable, and services cannot be developed or improved [45]. Researching sensitive issues with vulnerable people has in fact rarely been found to cause distress for participants [46]. In the small percentage of people for whom this has occurred, the distress is not thought to be long lasting or harmful [47] and does not prevent them from feeling that they have benefited from the experience [48, 49].

Perceptions of harm and burden were identified as affecting recruitment to this study. There are a number of studies involving extensive assessment that are linked to ARCOS IV, and the clinical researchers on the parent incidence study are the main point of contact for recruitment. Some of these researchers expressed that they initially experienced some guilt when asking people if they would also like to volunteer for the qualitative study because of the substantial time commitment required for participation in the parent incidence study. These concerns have been explored and addressed in a number of ways. A key component has been regular attendance of the qualitative researcher at team meetings. This enables updates and initial findings to be shared alongside education and discussion regarding the ethical issue of deciding for participants instead of giving them the opportunity to choose for themselves.

Other discussions highlighted the benefits of giving people the opportunity to have their story recorded to improve services in the future. This is seen as valuable because it provides a space for recording information that is not possible given the highly structured questionnaires and measures being used in the parent incidence study [50, 51]. Further, as a result of these discussions, sensitive issues training has been put in place to assist researchers to manage their own emotions when listening to troublesome stories [52-55].

Reducing burden on participants and maximizing benefits

Taking part in research should arguably be a valuable or beneficial experience for those taking part and steps should be taken to reduce any burden [56]. Interviews are therefore conducted at times and locations that suit the participant to reduce inconvenience as much as possible [56]. The key researchers on this study have backgrounds and experience in community and rehabilitation settings, enabling them to be sensitive and responsive to distress and fatigue for example, and onward referral is arranged if required. Participants are asked about what has gone well and interviews are concluded, wherever possible, on a positive note such as reflecting on improvements to date, support received or valued activities [57]. Training for the researchers is given where required to ensure appropriate levels of experience and expertise, and supervision occurs regularly [56]. The building of rapport is balanced with responsible questioning, using professional skill and discretion, knowing when to invite further detail and when to listen in silence so that participants tell the parts of their story that they want to and do not feel pressured to reveal too much [58].

Participants have frequently stated that one of their motivations for participating in the study is to ensure better services and experiences for others. This emphasizes researcher responsibility to ensure a methodologically robust study that addresses the identified gap in knowledge and contributes to improving services and experiences of those that follow [59]. This sort of methodology is thought to be particularly appropriate for the development and critique of clinical interventions [60]. In part this is due to data collection and analysis methods seeking information about how to best manage specific health issues directly from participants and describing this in language that people with the condition and clinicians who translate this knowledge into practice, will understand [30].

Ensuring diversity of experience

Despite the efforts of recruitment noted above, initially those taking part experienced relatively mild stroke and were predominately New Zealand European as compared with the more variable population recruited to the parent incidence study. Some participants with stroke who were registered for the parent incidence study were initially too unwell or had significant communication difficulties, which precluded their ability to take part in the assessments. On these occasions a proxy was nominated to take part on their behalf, necessarily precluding participation in our study with the original inclusion criteria. Once the lack of diversity with regard to residual disability and ethnicity within the qualitative study sample were noted, two steps were taken. First further participants were recruited from the parent study targeting ethnicity and stroke severity. The second step was to seek ethical approval to invite proxy whānau/family members to take part in interviews to ensure findings reflect this important group's experience. Interviews involving whānau/family member and proxies seek information about what it is like to support a person with stroke, acknowledging that stroke affects both the individual and the people close to them. The sample recruited are being regularly reviewed in light of early findings in order to identify whether further steps are required to ensure appropriate heterogeneity [36].

Facilitating participant retention

Retaining participants over the course of a longitudinal study has been identified as problematic [61, 62]. Learning what makes participation ‘worth it’ is important. Participants in our study have reported a number of benefits to participating in this, and the parent incidence study, particularly valuing being followed up and monitored and that important aspects of their recovery are being recorded given the paucity of community services for people with stroke. Participants also request information, take opportunities to ask questions about their stroke and recovery, and discuss concerns and issues. Conscious attention is given to the building of relationships [63]. On commencement of the interview, time is allowed for participants to sense the interviewer is truly hearing them rather than being there to complete ‘a task’. On completion of the interview, steps are taken to ensure interviewees are not distressed and have the researcher's contact details. We are also investigating other approaches to maintaining contact including multiple tracking techniques, six monthly study updates and maintaining flexibility around appointment times, and by being willing (and able) to travel long distances in order to complete interviews [63]. The aims and importance of the study are explained at every contact point, acknowledging the participants' contribution to these.

Substudy on self-management

Interviews include a targeted focus on self-management within participant and whānau/family member interviews, which will contribute to a PhD being completed by the clinical researcher working on this study (S. R.). From the interviews, data relating to strategies for reengaging in everyday life and self-management skills will be sought and specifically analyzed. The aim is to gain a better understanding of the clinical issue of self-management after stroke in order to contribute to best practice in this area.

Remaining grounded – use of a community reference group (CRG)

We have formed a CRG in order to inform, enrich, and support the study. This is comprised of a small group of people representing people with stroke and providers who meet quarterly with the operational working group. The primary aims of the CRG are to consider the translation of research from the beginning of the study, to advise the team how to better connect the study with the community and ensure that the study reflects and addresses the needs of people with stroke and their whānau/family and develops with appropriate consultation. This group also comments on design, methods, analysis, study progress, findings, and dissemination. The potential to involve this group in the planning of future studies is also being explored. Indeed, our emphasis on truly ‘being present’ during interviews was highlighted in the CRG as being fundamental if we wished participants to tell us their stories.

Added value of the qualitative component to ARCOS IV

While quantitative research necessarily involves some qualitative interpretation and judgment of data and analyses [64] it is becoming much more common for clinical research to include a ‘formal’ qualitative component (thereby introducing mixed methods) [64-66]. The wide range of approaches to mixed methods has been well summarized elsewhere [67] with the most common approach being to combine structured interviews and questionnaires within a cross-sectional design [68]. Our study should advance mixed methods research in stroke, being longitudinal and having a larger and more diverse sample than many other qualitative studies in stroke. By explicitly linking to the longitudinal population-based quantitative study, our intention is to: enable a richer understanding of the ‘experience’ of recovery and adaptation; address factors that are not amenable to the standardized measurement; promote new questions of importance to people with stroke for further research; and augment consideration of surprising or unexpected findings in the quantitative study.

There are certainly a number of problematic issues that arise in mixed methods (most commonly that one or other approach is an afterthought or ‘weak’ add-on) [69-77]. Our approach, ensuring clear communication within and between the two study teams, shared core staff and joint meetings should enable us to provide novel findings, and methodological advances of how other epidemiological research might utilize qualitative methods.

Summary

  1. Top of page
  2. Abstract
  3. Introduction
  4. Objectives of this paper
  5. Methods
  6. Summary
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

This study offers two specific benefits to what we know about stroke. First, the study is one of few qualitative investigations in stroke to benefit from sampling from a representative population of stroke survivors. In addition, longitudinal qualitative studies have rarely been conducted in stroke and the information on changing perspectives after stroke will increase knowledge of the natural course of recovery and provide information on access to and satisfaction with stroke services, especially for Māori and Pacifica people who are at high risk of stroke and face inequities in health-care accessing; and identify service gaps/unmet needs to ensuring evidence-based policy, resource allocation, prevention planning, management services, and evaluation of service performance.

There are also some specific benefits to running quantitative and qualitative studies concurrently. Communication and discussions between the two teams have assisted with addressing some of the methodological challenges that have arisen to date.

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Objectives of this paper
  5. Methods
  6. Summary
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

This study aims to identify the key barriers and facilitators to recovery and adaptation as perceived by stroke survivors over the first three-years following stroke to inform support interventions for New Zealand stroke survivors and their whānau/families. We have experienced a number of challenges as we attempt to achieve the study aims. However, what we have learned as we have developed strategies to address these difficulties is potentially useful for other teams attempting to respond to the increasing interest in combining qualitative research with population-based studies. We would highlight the importance of communication between research teams working on linked studies, the sharing of early findings, ongoing and responsive researcher training, consultation with a CRG and a plan for managing large amounts of data prior to collection and analysis.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Objectives of this paper
  5. Methods
  6. Summary
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

The authors wish to acknowledge the contribution of the study participants, the ARCOS IV research assistants, the medical staff at all the Auckland public hospitals, and the members of our CRG.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Objectives of this paper
  5. Methods
  6. Summary
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Objectives of this paper
  5. Methods
  6. Summary
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information
FilenameFormatSizeDescription
ijs12164-sup-0001-si.doc36K

Appendix S1. Interview guide.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.