A secondary analysis was used to address the objective of this study. The process for secondary analysis described by Johnston [25] served as a guide. The main steps in the process were (1) Develop research questions, (2) Identifying datasets, and (3) Evaluating datasets to ensure the appropriateness for the research topic.

Data for the participants in Sweden were extracted from a previous study conducted by the last author [26], and the objective of the study was to describe participation in activities and explore the relationship with SHCs among persons aging with a traumatic SCI. A total of 121 community-dwelling persons registered at the regional SCI outpatient center, among those, a total of 73 answered the survey by phone and thus constitute the Sweden sample.

Data for the USA participants were extracted from a larger national sample with 604 participants (of which 139 had an SCI), who were recruited through several national organizations working for people with disabilities. Data collection was done through a web-based survey. Data from this sample have been used for several articles with different objectives [14, 27] Inclusion and exclusion criteria for the Swedish original study and the larger national sample in the USA are shown in Fig. 1.

*According to the ASIA Impairment Scale (ASIA) [31]. **SCI, multiple sclerosis, cerebral palsy, etc.

To address the objective of this current study, a purposive sampling was used within the USA data, to identify a study sample matched with the Sweden study, based on the following inclusion criteria: (1) men and women with an SCI, (2) aged 50 years or older, and (3) at least 20 years post-injury. This resulted in a USA sample of 45 participants for the current study.

The Participation Survey—Mobility version 3 (PARTS-Mv3) is a self-report instrument designed to measure participation of persons with mobility impairments in different activities. An advantage of this instrument is that it captures several subjective dimensions of participation and provides information about what limits participation.

It is divided into two sections; the first one, referred to as the Characteristics of the Respondent (CORE), measures items such as general health on a 5-point Likert scale, with higher scores indicating better health, the prevalence of SHCs, and the general use of assistive devices. The second section measures participation for persons with mobility impairments in 24 activities. For each activity, four components of participation (temporal, evaluative, health-related, and supportive) are measured. Temporal questions focus on frequency of, or time spent participating in an activity. Evaluative questions focus on choice, control, satisfaction, and importance of participating in a particular activity. Health-related questions focus on whether the activity is limited by for example pain and/or fatigue. Supportive questions address the personal and/or assistive devices that the person may use to participate. The original PARTS has been tested as a reliable measure with moderate-to-high internal consistency and stability [28]. In the USA version of the survey, the University’s Institution Review Board requested a response option for individuals who preferred not to answer the questions. In the Sweden version, individuals only answered the questions for the activities in which they participated.

To answer the objective for the study, the researchers decided to use the temporal, supportive, and the health-related questions for 10 of the 24 PARTS activities. The selection of the 10 activities was based on the findings in an earlier study by the last author [16]. Data analyses were conducted using the IBM Statistical Package for Social Sciences (SPSS) statistics, version 22.0 (Armonk, NY: SPSS, Inc.).

To guide the data analysis, the following questions were used: (1) Do differences in general health and/or SHCs exist between persons with a long-term SCI in Sweden and the USA? (2) Are there differences in the use of assistive devices and personal assistants between persons with a long-term SCI in Sweden and the USA? (3) Are there differences in participation in activities between persons with a long-term SCI in Sweden and the USA? Descriptive statistical analyses (e.g., mean, standard deviation, and frequencies) were performed to explore and describe the data and the study population. To examine differences between factors such as age, time since injury, SHCs, and participation in activities, the independent t-test, crosstabs, and X2 test were used where appropriate to compare the samples. The significance level was set to P < 0.05.

This study expands on two separate studies that were conducted with approval from the Human Ethics committee in Stockholm, Sweden (no: 2014/214-31/3) and the University’s Institution Review Board, USA (no: 201101785), and where informed consent was obtained from the participants. We certify that all applicable institutional and governmental regulations concerning ethics were followed during course of this research.