Background
Complex interventions are challenging to evaluate since they contain components that may act both independently of each other and interdependently, which makes it complicated to assess individual aspects of the intervention. The challenges are, among others, related to difficulties in standardising the study design and delivery of the intervention, and assessing the impact of local contextual factors [
1]. The British Medical Research Council (MRC) [
1] has developed a research framework for complex interventions. They suggest a multi-step approach, including a development phase, followed by feasibility/piloting, evaluation (preferably through a randomised controlled trial) and implementation. They conclude that evaluation of complex interventions requires a mix of both quantitative and qualitative methods to get a more comprehensive understanding of the interventions. One challenge in complex interventions is the length and complexity of causal chains linking the intervention with outcome [
1]. This means that the intervention may affect important factors that were not planned for and not measured with quantitative methods [
1], factors that could be discovered with qualitative methods. Thus, knowledge from qualitative research of a complex intervention is crucial to fully interpret the results of and understand an intervention.
Case management is a complex intervention that has been used in different health care settings such as psychiatry and geriatric care [
2,
3]. It has no single definition but it has been suggested that basic case management may include identification and outreach, comprehensive individual-based assessment, care planning, care coordination, service provision, monitoring, evaluation and meeting individual needs [
4,
5]. Several studies have investigated the effects of case management for older people. These studies mainly focused on outcomes such as healthcare utilization and costs, quality of life, physical or cognitive functioning, quality of care and patient satisfaction [
6]. The reported effects are contradictory [
6]. In addition, such studies are seldom described in detail [
7], which makes it difficult to compare them. Thus, there is a need for in-depth investigations to gain deeper understanding of the interventions, which will allow comparisons to be made and will enable us to draw conclusions about best practices [
8]. To be able to further develop case management interventions a greater understanding of the intervention’s content and construction is needed. Qualitative studies are important to identify different barriers and facilitators that could be underlying reasons for an intervention being successful or not [
9] and are necessary for implementation [
10]. A qualitative evaluation is necessary to obtain a comprehensive description of the intervention’s components, to explore conditions for implementation, to establish construct validity and to facilitate possible replication [
9]. Furthermore, it is essential to investigate different perspectives of the experience of an intervention, from both those receiving and those performing the intervention. However, studies of these different perspectives are generally lacking.
Some qualitative studies have focused on experiences of case management for older people [
11‐
13]. The studies by JM Nelson and P Arnold-Powers [
12] and K Brown, K Stainer, J Stewart, R Clacy and S Parker [
11] found that the relationship between the case managers (CM) and participants was highly valued. JM Nelson and P Arnold-Powers [
12] reported that the relationship with CM helped to provide security, safety and comfort for clients, and K Brown, K Stainer, J Stewart, R Clacy and S Parker [
11] reported that their participants experienced that the CM had improved their quality of life. P Sargent, S Pickard, R Sheaff and R Boaden [
13] and [
11] also found that the participants were satisfied with the CMs’ different skills and ability to arrange services. According to P Sargent, S Pickard, R Sheaff and R Boaden [
13] psychosocial support was emphasised by both the patients and carers with experience of case management, and was viewed as being equally important as clinical care. Complex interventions are dependent on the local context [
1], which means that case management interventions could be experienced in various ways and have unique problems depending on the context they are performed in. Thus, each intervention needs to be explored in terms of what has been done and also how it was experienced, from both the providers’ and the receivers’ perspectives.
The aim was to explore older people's and case managers’ (CM) experiences of a complex case management intervention.
Method
The study had a qualitative design, using opened-ended interviews with older people who were part of a case management intervention and the CMs who had performed the intervention.
Participants
The study comprised 20 people: 14 participants (four men and 10 women, age 75-95 years, median age 83) who had received the case management intervention and six CMs (four nurses and two physiotherapists, age 31-51 years, median age 44) who had performed the intervention. Inclusion criteria for the older persons were: (1) age at least 65 years, (2) residence in an ordinary home, (3) need for help with two or more activities of daily living (self-reported and meaning that the participant could not perform the whole activity by them self, for example cleaning, transportation, and or managing medications), and (4) admission to hospital at least twice, or at least four visits to outpatient care, in the 12 months prior to entering the intervention study. In addition, participants had to be cognitively adequate and feel well enough to participate in an interview. Cognitive status was examined by using the Mini Mental State Examination (MMSE) [
14]. The instrument covers cognitive areas of orientation, memory, attention, the ability to name, the ability to follow verbal and written commands, write a sentence spontaneously, and copy a complex polygon. Generally accepted cut-off points are; 25-30 for normal cognition; 21-24 for mild cognitive impairment; 14 or below for moderate or severe cognitive impairment [
15]. In this study a cut-off of 25 points or higher out of a maximum of 30 required for participation. All participants included in the CM intervention had been recruited from a nearby university hospital, from the four primary care centres in the study municipality, through the municipal home care organisation or by the participants contacting the research group by themselves. The participants in the present study were recruited face-to-face by the research team during their participation in the CM intervention study. Purposeful selection [
16] was used to obtain variation in gender and in age, use of home care services and CM (Table
1). Before entering the case management intervention study, the participants were informed, both in writing and verbally, that they might be asked about being interviewed with open-ended questions. This information was then repeated after nine months of the one-year intervention, when they were asked to participate in the interview.
Table 1
Characteristics of the interviewed receivers and providers of the intervention
R0005 | Male | Widower | 12 months | No |
R0012 | Female | Widow | 13 months | No |
R0020 | Female | Widow | 13 months | No |
R0025 | Female | Widow | 13 months | No |
R0026 | Female | Divorced | 14 months | Yes |
R0029 | Male | Living apart | 14 months | Yes |
R0031 | Female | Widow | 12 months | Yes |
R0036 | Male | Widower | 12 months | No |
R0053 | Female | Married | 15 months | No |
R0079 | Female | Widow | 18 months | Yes |
R0081 | Female | Widow, living apart from a new man | 16 months | Yes |
R0083 | Male | Widow | 15 months | Yes |
R0091 | Female | Unmarried | 13 months | No |
R0143 | Female | Married | 12 months | No |
Providers
|
Code
|
Profession of the CM
|
Code of the participant
|
Gender of the participant
|
Participant receiving home care services
|
N0004 | CM1 - Nurse | R0004 | Female | No |
N0025 | CM2 - Nurse | R0025 | Femalez | No |
N0083 | CM2 - Nurse | R0083 | Male | Yes |
N0055 | CM3 - Nurse | R0055 | Male | No |
N0114 | CM3 - Nurse | R0114 | Male | No |
N0086 | CM4 - Nurse | R0086 | Male | No |
N0161 | CM4 - Nurse | R0161 | Female | No |
P0028 | CM5 - Physiotherapist | R0028 | Female | Yes |
P0081 | CM5 - Physiotherapist | R0081 | Female | Yes |
P0085 | CM5 - Physiotherapist | R0085 | Male | No |
P0095 | CM5 - Physiotherapist | R0095 | Female | Yes |
P0098 | CM6 - Physiotherapist | R0098 | Male | No |
P0134 | CM6 - Physiotherapist | R0134 | Female | No |
P0151 | CM6 - Physiotherapist | R0151 | Male | Yes |
P0169 | CM6 - Physiotherapist | R0169 | Female | No |
Interviews were also conducted with the CMs. Six CMs were interviewed, two of whom had been educated as physiotherapists and four as nurses. The CMs were, depending on the number of participants, employed on a part-time basis in the research project. They were all recruited from municipal, primary care or hospital settings. The CMs were interviewed about every individual that they had met. In total, 162 interviews were made. Purposeful selection of fifteen interviews was used to obtain variation in CMs, gender, age and use of home care services. The CMs worked in the research project for between 2 and 5 years and had experience of caring for or rehablitating older people (Table
1).
Setting
The case management intervention was in addition to standard care and the participants were consecutively recruited between 2006 and 2011. The intervention, conducted in the southern Sweden, was a one-year home-based case management intervention with home visits at least once a month [
17]. The intervention comprised four components:
traditional case management (including assessment, care planning, follow-up, care coordination, home visits, telephone calls and advocacy),
general information (about the healthcare system, social activities, nutrition and exercise, among other things),
specific information (related to the respondent’s specific health status, individual needs and medication) and
safety and continuity (availability of CM by cell phone during working hours) [
17].
The CM study was developed according to the MRC’s framework for complex interventions [
18]. The pilot study phase, in which the intervention was developed, is described elsewhere [
17]. Changes after the pilot study have also been reported [
19].
Data collection
Data were collected by means of personal interviews. The interviews were conducted between 2007 and 2012, and were conducted by four different persons due to a change in staff during this period. The first author (M.S.), the fourth author (J.K.), and two research assistants (one male and one female) conducted eleven, eight, eight and two interviews, respectively. The interviews were semi-structured, which meant that they were neither fully structured nor fully unstructured. The participants were free to talk about any subject, but the interviewer guided the interview [
20]. Two thematic interview guides were used – one for the participants and one for the CMs – to ensure that the interviews covered the same areas of content. The CM interviews covered two themes: (1) the person they met and how the contact started, what they had done and what effects they thought this might have had; and (2) how they perceived the intervention, whether there was something that they considered successful or unsuccessful. The interview guide for the participants did not only comprise questions about the intervention: as well as questions on “help and support” (including questions about the CM and the case management intervention), it also covered “health”, “contacts with the healthcare system” and “the future and concerns”. Open questions were used and included questions such as “could you tell me about an ordinary meeting with the case manager?” (to the participant) and “could you tell me about this person that you have met in your role as case manager?” (to the case manager). Probing questions could for instance be “could you give an example?”, “how did that feel?” and “What did you do then?”. The interview guides were changed slightly during the study meaning that the order of the questions where changed, and thus all interviews covered the same areas. All interview guides were tested in pilot interviews on both the participants and CMs. No major changes were made in the interview guides after the pilot interviews and thus included in the study. Each interview started with clarification of the aim of the interview and the interviewee’s right to terminate the interview whenever he/she wanted.
The interviews with participants were conducted after they had received the intervention for at least nine months in order that they had undergone the majority of the intervention. They were interviewed after a mean of 14 months after they were included in the CM intervention. The interviews were carried out in a place chosen by the participant. All interviews took place in the participants’ homes and were between 40 minutes and 2 hours 51 minutes long. During the interviews, no-one besides the participant and the interviewer was present. However, in one interview the sister was in an adjacent room and the participant asked her some questions.
Interviews with the CM were made for each participant they had met after the participant had received the intervention for at least nine months. The CM interviews were conducted after in mean 17 months after the participant had been included in the intervention study. The CMs had with them the case records of their participants. All CM interviews took place at the department of the researchers and lasted between 9 and 24 minutes. All interviews were audio recorded and transcribed verbatim.
Analysis
The interviews were analysed by content analysis. The analysis was influenced by B Berg [
21], who suggests that content analysis may comprise a combination of both manifest and latent analysis. The manifest part concerns what is said and is visible in the text, while the latent part concerns finding an interpretable structure, a deeper underlying meaning [
21]. The analysis was made using different steps inspired by UH Graneheim and B Lundman [
22]. In the first step, the transcribed interviews were read several times independently by all authors to obtain a sense of the whole. In the second step, meaning units related to the aim were identified from the text. The third step involved condensing the meaning units into codes. The next step embraced a movement between the meaning unit and the text, between the text as a whole and its parts. During this process, subcategories and categories were identified. Three interviews were analysed independently by the first (M.S.) and last (J.K.) authors. The subcategories were then discussed by M.S. and J.K. until a consensus was reached. The first author then analysed an additional number of interviews and the subcategories were again discussed by M.S. and J.K. Groups of subcategories sharing the same content were arranged under tentative categories. The remaining interviews were divided between M.S. and J.K. and analysed independently. M.S. and J.K. discussed the content of the subcategories and developed categories. Finally, the four authors discussed the findings until a consensus was reached, and additional small adjustments were made to the categories. Quotations were chosen to illustrate the different subcategories. Examples of the analysis process are presented in Table
2.
Table 2
Examples of the analysis process
N0086 | He had not really the insight that there was something seriously… but he just laughed it off when you talked about it. | A failure to reach | Dealing with barriers | ENTERING A NEW PROFESSIONAL ROLE |
P0028 | We talked about it… about residential care for her. And if it… tried… well, talk a bit about what it was like to have some people around and so. But she… no. She did not want to. She did not want much at all [laughs a bit]. | Meeting people that do not want to be helped or do not want to incommode | | |
N0025 | You become despondent when you do not succeed. But … but I have offered it anyway. It sure is tough, so, it is. | To feel personal involvement | Setting limits | |
R0031 | Yes we're talking about everything … I think. I do not remember anything exactly… but we have… we are talking about everything. Yes it is just as if we have become friends. I see it as if she has become my friend (pause). | To feel confident in a person and her competence | Reliable competence | A POSSIBLE ADDITIONAL0020RESOURCE |
R0079 | And it's never in a hurry either, but they… There was never any hurry. Never ever. And they were helpful. | To get a chance to build a stable relationship | | |
R0143 | Well you, that (pause) that I have not needed to search for [health care] because all I have needed… uh to ask for uh I have uh used the case manager for that… | To find a replacer for the usual health system | Gaining a safety net | |
Ethical considerations
This study was approved by the Regional Ethical Review Board in Lund (Ref. nos. 342/2006 and 499/2008) and is registered at Clinicaltrails.gov (Ref. No NCT01829594). All participants provided written informed consent for participating. The participants’ autonomy was acknowledged by emphasising, both before and at the beginning of the interview, that participation was voluntarily, and that the participant could withdraw from the study at any stage. They also were also informed that confidentiality should be maintained when presenting the results.
Acknowledgements
This project was carried out in collaboration between the Faculty of Medicine at Lund University, the Swedish Institute for Health Sciences (Vårdalinstitutet), Skåne University Hospital, primary in Eslöv and Eslöv municipality. We are grateful to the Faculty of Medicine at Lund University, Vårdalinstitutet, Region Skåne, Johan and Greta Koch’s Foundation, the Swedish Association of Health Professionals (Vårdförbundet), the Swedish Society of Nursing and Södra Sveriges Sjuksköterskehem (SSSH) for funding this study. We are also most grateful to the participants and their next of kin. We would especially like to thank the following people: Sara Modig, physician at Tåbelund Primary Care Centre; head nurse Magdalena Andersson of Eslöv municipality; and registered nurses Marie Louise Olofsson, Jeanette Hellberg, Lena Jönsson and Jenny Linderstål, and registered physiotherapists Caroline Larsson and Ulrika Olsson Möller, all of whom worked as CMs. We would also like to thank Stephen Gilliver for revising the English in this article and for translating the excerpts from the interviews.
Competing interests
The authors declared that they have no competing interests.
Authors’ contributions
MS participated in the design of the study, in the collection of data, performed the analysis and interpretation of data, and drafted the manuscript. PM participated in the design of the study and helped in the analysis of the data and to draft the manuscript. UJ participated in the design of the study, participated in the design of the study and helped in the analysis of the data and to draft the manuscript. JK participated in the design of the study, participated in the data collection, performed the analysis and interpretation of data, and helped in the drafting of the manuscript. All authors read and approved the final manuscript.