Study design and Setting
This qualitative study was conducted as part of the LiFE-is-LiFE trial (19), a 12-month multi-center, single-blinded, randomized non-inferiority trial comparing gLiFE with LiFE regarding fall reduction and cost-effectiveness (ClinicalTrials.gov, NCT03462654) (period: 06/2018 to 08/2020). Participants with a verified fall risk took part in either seven gLiFE or LiFE sessions within eleven weeks, followed by two phone calls. Follow up assessments were performed after six and twelve months.
For the present study, qualitative data collected in four focus group (FG) discussions after the six months follow up assessments (04/2019) were analyzed.
Programs
LiFE aims to reduce fall-related outcomes and promote long-term physical activity in community-dwelling older adults by integrating balance and strength activities into daily routines. gLiFE (9) and LiFE (20) are described elsewhere. Table 1 provides an overview of similarities and differences of both formats.
Table 1
Similarities and differences between LiFE and gLiFE conducted in the LiFE-is-LiFE trial (19)(19)
| LiFE | gLiFE |
Brief aim | Improve balance and lower limb strength, increase physical activity, decrease risk of falling, long-term sustainability of the LiFE activities through habit formation and self-empowerment |
What: Materials | Participant’s manual, German version (22); Contains descriptions and instructions of LiFE activities; principles of balance and strength training as well as physical activity enhancement; safety instructions when performing the activities; background on balance and strength exercise; assistance and support for changing habits and performing LiFE activities Trainer’s manual, German version; one for LiFE, one for gLiFE. Contains all information also included in the participant’s manual; additionally: outline of all 7 sessions and 2 phone calls, including text templates, material, preparations, and precautions Workbook; for all participants; used during intervention: Includes information on study procedures, personnel, contacts, and safety instructions; activity planning sheets for balance, strength, and physical activity; activity counter, notes pages; LiFE principles Aids and materials during intervention sessions: Laminated cards, showing LiFE principles and LiFE activities to be used as visual aids during intervention sessions; balls, blankets, sponge rubber, boxes, clipboards, pens, bags, name tags, flipcharts |
What: Procedures | 7 home visits by one qualified trainer; 2 phone calls 4 and 10 weeks after last session | 7 group sessions (n = 8–12 participants) led by one main and one co-trainer, 2 phone calls 4 and 10 weeks after last session |
Who provided | Trainers are sport scientists, physiotherapists, occupational therapists or psycholo-gists. All trainers received a two-day training course on the program background, aims, and components prior to the project start. |
How | One-to-one situation in the participant’s home | Group setting with 8–12 participants and two trainers |
Where | Two study sites: Heidelberg and Stuttgart (Germany) |
When and how much | 7 sessions within 11 weeks: week 1, 2, 3, 5, 7, 9, 11. Two phone calls 4 and 10 weeks after the last session (i.e. week 15 and 21). Duration of each session: 1-1.5 h | See LiFE Duration of each session: 2-2.5 h |
Setting | Intensity and dose are determined by the individuals’ activity plans, adherence, and performance level of each activity |
Behavior change | Behavior change theories based on LiFE trainer’s manual and participant’s manual (20). | Modification of the original behavior change concept using established theories on health behavior, such as the Health Action Process Approach (48, 49) and the Self-Determination Theory (39). Intervention contents of gLiFE were mapped using the Behavior Change Technique (BCT) Taxonomy v1 (50). |
(please INSERT Table 1 here)
Trainers teach the participants how to perform the activities (e.g., squat), where (e.g., in the bedroom), and when to implement these into daily routines (e.g., each time when reaching for the floor drawer). New movement habits are created by linking the LiFE activities to specific daily situations based on behavior change concepts (9, 21).
Participants learn how to adapt chosen activities to their lifestyle and how to increase difficulty to ensure progress using LiFE principles (20).
The main difference between programs is the delivery format: group delivery for eight to twelve participants by two trainers (gLiFE) compared to one-to-one delivery in participants’ homes (LiFE). In gLiFE, the trainer’s role is to teach and facilitate; in LiFE, the trainer teaches and substitutes a training partner. Contents of gLiFE and LiFE are taught in predefined order, but teaching in gLiFE is organized in an interactive manner including group discussions and joint activity practice with peers.
All participants receive the German participant’s manual (22) and a workbook, including a modified activity planner (9), and an activity counter to plan and monitor activity performance.
Participants
310 community-dwelling older adults (> 70) were randomized to either gLiFE or LiFE at two study centres in Germany (Network Aging Research, Heidelberg University; Robert-Bosch-Hospital, Stuttgart). For this study, 30 participants (22 women, 8 men; Mage=78.8; range 70–96 years; ngLiFE=15; nLiFE=15) were purposively selected (23) from the trial. We included 15 participants (nwomen=13, nmen=2) who adhered to the program after completion, and 15 participants (nwomen=9, nmen=6) who indicated lower adherence. Habit strength was used as an indicator for intervention adherence and assessed by the Self-Reported Behavioural Automaticity Index (SRBAI) (24). The SRBAI was measured at 6-month-follow up to identify whether LiFE activities have become habitual. The SRBAI median split (25) was defined as the threshold (SRBAI ≥ 4.49 = higher behavioral automaticity, SRBAI ≤ 4.49 = lower behavioral automaticity). Each FG included participants showing higher and lower behavioral automaticity, different ages, and gender to maximize the breadth of information and foster discussion between participants that did and did not successfully implement LiFE. Participant characteristics are summarized in Table 2; they were predominantly female (73%), highly educated, at risk of falling indicated by the Timed Up-and-Go Test (26), and were cognitively healthy according to Montreal Cognitive Assessment (MoCA) (27). Participants were invited via telephone. The recruitment process is shown in Fig. 1 (see Fig. 1).
Table 2
Characteristics of participants (N = 30)
| Mean (SD) or % (n) |
| Total (N = 30) | gLiFE (N = 15) | LiFE (N = 15) |
Age, years | 78.8 (6.6) | 78.5 (6.1) | 79.1 (7.2) |
Women | 73.3 (22) | 73.3 (11) | 73.3 (11) |
MoCA, score (0–30) | 26.0 (1.7) | 26.5 (1.8) | 25.4 (1.5) |
Fall incident baseline | 30 (9) | 20 (3) | 40 (6) |
Completed years of education | 14.1 (3.8) | 14.5 (4.5) | 13.7 (3.0) |
TUG, sec. | 12.7 (3.4) | 12.7 (2.7) | 12.7 (4.0) |
Note. MoCA = Montral Cognitive Assessment (27); Fall incident = falls in the last six months at baseline measurement; TUG = Timed Up-and-Go Test (26). |
(please INSERT Fig. 1 here)
Data Collection
An interdisciplinary team of exercise scientists, psychologists, and physiotherapists developed a semi-structured interview guide (see Additional file 1) for both formats based on a previous LiFE FG discussion (9).
The interview guide comprised 14 main questions on the programs and their components, the group and individual format, and habit formation processes like action planning or building new movement habits. It was piloted with one LiFE-is-LiFE participant regarding clarity of questions and refined after pilot evaluation.
Two FG discussions were conducted at seminar rooms in each study center, one for gLiFE (nStuttgart=8, nHeidelberg=7), one for LiFE (nStuttgart=7, nHeidelberg=8) lasting between 90–100 minutes. At the beginning, the study purpose (evaluation of program acceptability) was explained, and participants gave written informed consent. The moderator facilitated discussions by asking questions with follow-up prompts, probing, encouraging reserved participants to speak, and ensuring that discussions covered the main topics. The moderator (main author, physiotherapist, and external researcher) and co-moderators (team members) were not involved in follow-up assessments and intervention delivery. Co-moderators took notes and kept time. Two of the co-moderators who took part in the program development stayed silent during discussions. FGs were audio recorded and transcribed verbatim in German, according to transcription guidelines by Kuckartz (28).
Data Analysis
Qualitative content analysis according to Mayring (29) was performed using inductive category formation. Coding was managed using NVivo 12 (QRS International, Australia). To explore acceptability, inductive categories were formed to identify participants’ experiences with program features like content, delivery, and implementation of both LiFE formats. All FG discussions were defined as the unit of analysis and the manifest content was analyzed. First, the main author formulated a category definition as a selection criterion to determine the relevant material from the text, and a level of abstraction, which defines how general or specific categories must be formulated (29). Based on this, text were coded line-by-line and a category was constructed every time an element of the text matched the category definition (see Table 3). The interview guide shaped the formation of initial categories. After 50% of the text, categories and coding rules were revised, then two authors (LR, FK) independently coded the text. Main categories were formulated and discussed. In case of disagreement a third researcher (SL) was consulted. Finally, categories were organized into overarching themes and contents were contrasted by group (gLiFE vs LiFE). Three authors (LR, FK, SL) agreed on the final category framework.
Table 3
Example of inductive category formation
Coding unit | Keywords | Category | Main category |
Group is more intense | Motivation because of group | Format of the program |
Group pressure is helpful |