Aging

Quality of Life and Community Reintegration

In the general population, advancing into older adulthood is a period when individuals are faced with a unique array of physical, functional, and environmental stressors. This is no different for individuals aging with a traumatic SCI, who are now living an average of 30 to 40 years post-injury (YPI) (Samsa et al. 1993). As more individuals with SCI survive into their second, third, and even later decades, living with a disability becomes a life-long process for persons with SCI (Hallin et al. 2000).

Given the evidence in the previous sections of this chapter indicating that SCI represents a model for premature aging in some body systems (e.g. cardiovascular and endocrine, musculoskeletal, immune, and respiratory systems), the physical and functional declines associated with natural aging are likely to present more quickly among individuals with SCI. Such knowledge of these effects of aging however is insufficient for rehabilitation purposes without any indication of how individuals perceive the aging-related changes and how they adapt their lifestyles in response to such changes (Charlifue et al. 2010).

A key goal of rehabilitation is to enable successful community reintegration and high QoL. QoL describes the well-being and life satisfaction of an individual, and is a multi-factorial construct, which includes but is not limited to self-assessments of interpersonal relationships and social support, physical and mental health, environmental comfort, and a host of psycho-social factors (Kaplan & Erickson 2000). Community reintegration is an important construct shown to be predictive of life satisfaction in persons with SCI (e.g. Pierce et al. 1999; Richards et al. 1999; Putzke et al. 2002b; Tonack et al. 2008; Kemp & Bateham 2010). Community reintegration has been defined as returning to family and community life, engaging in normal roles and responsibilities, and actively contributing to one’s social groups and to society as a whole (Dijkers 1998). Thus successful reintegration involves resuming occupations or activities deemed important to the individual and society (i.e. self-care, employment, leisure, etc.; Yasui & Berven 2009). Environmental factors (e.g. social, institutional, cultural or physical) can either create barriers or facilitate reintegration, which impacts QoL (Anderson 2004).

In the general population, older adults frequently experience physical declines (Branch & Jette 1983) that may limit their activities of daily living (e.g., Hoyer et al. 1999), and negatively impact community reintegration and QoL. Similarly, both physical and mental health factors influence QoL in persons with SCI. For instance, poor physical health, secondary health conditions (e.g. pressure ulcers, pain, etc.), depression and stress have all been shown to negatively affect QoL (Craven et al. 2012).

With regards to aging, however, there are some mixed findings in relation to community reintegration and QoL, even within the same studies. Some studies reports that life satisfaction, QoL and community reintegration (at least in some domains) improve with years post-SCI (e.g. Zarb et al. 1990; Pentland et al. 1995; Westgren & Levi 1998; Dijkers 1999; Tonack et al. 2008), whereas other studies indicate older age is associated with poorer community reintegration and QoL (e.g. Crewe & Krause 1990; Eisenberg & Saltz 1991; Whiteneck et al. 1992; Tonack et al. 2008). The discrepancies with aging and QoL tend to be more evident in cross-sectional analyses whereas longitudinal studies “mostly show relatively high and stable levels of QoL over long periods of time” (Kemp & Ettelson 2001, p. 119; Savic et al. 2010). An additional point to consider is that these differences may arise due to the use of different instruments, which may not all assess the same underlying QoL construct.

In this section, one systematic review, twenty-five longitudinal studies and two cross-sectional studies on community reintegration and QoL after SCI are reviewed.

Table 14. Systematic Review on Quality of Life and Community Reintegration

Author Year; Country Date included in the review Total Sample Size Level of Evidence Type of study Score	Methods	Outcome
Sakakibara et al. 2012; Canada Reviewed published articles from 1980 to 2011 N=21 Level of Evidence: Modified Sackett scale (all were level 4) Type of study: Longitudinal AMSTAR: 7	Methods: Electronic databases were searched for peer-reviewed published studies reporting on age-related QoL changes over time in adults with traumatic SCI; data from relevant studies were transcribed into data extraction forms and analyzed by years post injury (YPI) and chronologic age. No interventions studied. Outcome measures: Include various QoL measures e.g., Satisfaction with Life Scale (SWLS), Quality of Life Index (QLI), Life Satisfaction Questionnaire (LSQ). Databases: MEDLINE/PubMed, CINAHL, EMBASE and PsycINFO.	In individuals with advanced YPI, overall QoL is consistently reported as good or excellent over time, however, with variations in different QoL domains. In 4 studies with samples with mean ages in the late teens and 20s and 5 or less YPI, both overall QoL and various domains of QoL were shown to significantly increase with age. 7 studies found individuals first assessed in their 20s (YPI ranging from 6-15 yrs) improved employment after 25 yrs of follow-up, but decreased satisfaction with social and sex lives, and general health. Similarly for individuals first assessed in their 30s, employment satisfaction increased but satisfaction with social and sex life, family relationships, recreational and life opportunities, emotional adjustment and control over life all diminished after a 9 year follow-up. 3 studies reported on individuals with at least 16 YPI and showed that although life satisfaction was variable over the yrs, after 16 yrs the reported life satisfaction was higher than at earlier follow-ups.

Table 15. Quality of Life and Community Reintegration

Author Year; Country Score Research Design Total Sample Size	Methods	Outcome
Celik et al. 2017; Turkey Level 2 evidence Longitudinal N=40	Population: 20 people with SCI in Group 1 (13 M, 7 F): mean age (y)=23.9±5.83 (range 18-44), mean YPI=9.35±6.06 (range 3-29), mean age at injury (y)= 14.6±2.8 20 people with SCI in Group 2 (10 M, 10 F): mean age (y)=43.85±12.77 (range 30-79), mean YPI=8.7±7.95 (range 1-31), mean age at injury (y)=33.1±16.6 Methodology: Depending on whether participants sustained their SCI <18 years old or >18 years old, they were categorized into Group 1 and Group 2, respectively, to investigate the factors that affect disability and QoL in people who sustained their SCI either before or after 18 years of age. Outcome Measures: Craig Handicap Assessment and Reporting Technique (CHART), World Health Organization Quality of Life Scale Short Form (WHOQOL-Bref), Beck Depression Inventory (BDI)	QoL scores were significantly higher for those that sustained their SCI after 18 years old. No differences between the 2 groups were found on BDI and CHART scores. No significant correlation was found between age of SCI onset, disease duration, ASIA scores, depression scores, total CHART scores for all participants.
Jorgensen et al. 2017; Sweden Level 5 Cross-sectional N=119	Population: 84 males and 35 females with SCI; mean age (y)=63.5±8.7 (range 50-89); mean TSI (y)=24±11.7 (range 10-56); Injury level and severity: Tetraplegia AIS A-C (18%), paraplegia AIS A-C (33%), all AIS D (50%); Working full/part-time (36%), disability pension/old-age pension (64%) Methodology: Data was collected from the larger Swedish Aging with Spinal Cord Injury Study longitudinal cohort survey assessing individuals ≥50 years old and ≥10 YPI. Participants were interviewed to describe all activities performed during the past 3 days to study the participation in leisure time physical activity (LTPA) among older adults with long-term SCI. Outcome Measures: Physical activity recall assessment to determine LTPA, and (bowel-related and bladder related problems, spasticity and pain)	Age and wheelchair use were significantly and negatively associated with total LTPA; 29% reported no LTPA, whereas 53% performed moderate-to-heavy intensity LTPA.
Krause et al. 2016; USA Level 2 Longitudinal N=49	Population: 49 persons with SCI (85.7% M) at 40-year follow up; mean age (y)=64.9±5.0, mean YPI=45.9±4.1; 55.1% had cervical injuries, 53.9% reported sensation below the level of lesion, and functionally, only 10.4% could walk. Participants were identified from outpatient records of a Midwestern USA university hospital in 1973. Methodology: Follow-ups were conducted in 1984 and at 10-year intervals thereafter to explore the natural course of aging after SCI by identifying changes in participation, employment, health and quality of life outcomes over 40 years. Outcome Measures: Life Situation Questionnaire (LSQ), employment rate, hours employed	Satisfaction with health, sex life and social life declined over time, whereas satisfaction with employment improved initially and was maintained over time. Self-rated current adjustment remained stable, whereas predicted future adjustment declined steadily over 40 years. Employment rate and average hours employed initially improved in the first two decades (49% to 71%) but decreased substantially during the last two decades of measurement (69% to 27%).
Lofvenmark et al. 2016; Sweden Level 2 Longitudinal N=38	Population: 27 participants with SCI attended yearly controls (YC) (67% M): mean age (y)=33±13; AIS A-C (78%), AIS D (22%); tetraplegia (44%), paraplegia (56%); Wheelchair user (81%), ambulatory (19%) 11 people with SCI did not attend the YC (82% M): mean age (Y)=29±13; AIS A-C (54%), AIS D (46%); tetraplegia (64%), paraplegia (46%); Wheelchair user (54%), ambulatory (46%) Methodology: All people who were admitted with a traumatic SCI during a 2-year period and survived to be discharged in a national SCI rehab center in Botswana were included as participants. Data was collected at the yearly controls (YCs) to describe the clinical and functional outcomes 2 years after traumatic SCI.	YCs had a higher rate of complete injuries, etiology of road traffic crashes and secondary complications during in-patient care, especially pain. Women rated ‘physical health’ significantly lower compared with men 59% experienced pain, 41% experienced UTIs, 48% experienced pressure ulcers, and 56% experienced spasticity but no deaths were reported One-third of the participants at follow-up had returned to work
Lundstrom et al. 2016; Sweden Level 5 Cross-sectional N=73	Population: 73 individuals with SCI (55 M, 18 F); mean age (y)=64.7±9.4 (range 50-87), mean TSI (y)=36.2±9.2; AIS A (56.2%), AIS B (6.8%), AIS C (16.4%), AIS D (20.5%); 20-35 YPI (n=33); 36-55 YPI (n=40) Methodology: Surveys were conducted by phone to collect information about the relationship between participation in activities and secondary health complications (SHCs) among persons aging with traumatic SCI. Outcome Measures: 10 activities from the PARTS/M-v3 (PARTicipation Survey/Mobility version-3), medical records	No statistically significant relations between participation in activities No statistically significant relations were found between the use of assistive devices and age or time since injury. For those who lived 36–55 YPI; increasing pain, fatigue, spasticity and decreased muscle strength were negatively affecting participation in activities, especially exercise and active recreation. A need to save strength/energy was cited as a reason for not participating in the activities.
Newman et al. 2016; USA Level 5 Cross-sectional N=768	Population: 768 people with SCI (71.9% M, 28.1% F); mean age at time of study (y)=54.3±11.9; mean TSI (y)=27±10.2; nonambulatory high tetra C1-C4 (8.9%), nonambulatory low tetra C5-C8 (28.1%), nonambulatory para (37.1%), ambulatory (25.9%) Methodology: The most recent data collected from the SCI Longitudinal Aging Study to evaluate the relationships between social disconnectedness and perceived isolation within an aging sample of participants with SCI through structural equation modeling. Outcome Measures: Life Situation Questionnaire (LSQ), Craig Handicap Assessment reporting Technique (CHART), Patient Reported Outcomes Measurement Information System (PROMIS), indicators of social isolation	Social disconnectedness was significantly associated with injury severity and chronologic age; perceived isolation was significantly associated with race; and both social disconnectedness and perceived isolation were significantly associated with years post-injury. Older participants were significantly more likely to report higher levels of disconnectedness. Greater time since injury was related to less objective and subjective social isolation after controlling for the effect of aging.
Pretz et al. 2016; USA Level 2 Longitudinal N=4846	Population: 4846 participants with SCI (78.1% M, 21.9% F); mean age at injury (y)=30±12; AIS D-E (19%), T1-S3 AIS A-C (41.5%), C5-C8 AIS A-C (25.0%), C1-C4 AIS A-C (11.1%) Methodology: The population sample was taken from the NSCID.. Data was collected during initial hospitalization and at post-injury years 1, 5, and 10, and every 5 years thereafter to describe individual trajectories of life satisfaction after SCI. Outcome Measures: Rasch-transformed Satisfaction With Life Scale (RTSWLS).	Initial life satisfaction was relatively low but steadily increased over time. Those who had higher initial life satisfaction still experienced increased life satisfaction over time but to a lesser extent than those who had lower initial life satisfaction scores.
Cao et al. 2015; USA Level 2 Longitudinal N=1032	Population: 1032 people with SCI (68.9% M);Age at injury (years)= 29.09±13.83 YPI= 13.27±9.21 Married or cohabiting (MC): n=409 (39.7%); Divorced, separated, or widowed (DSW): n=224 (21.8%); Single (S): n=397 (38.5%) Nonambulatory cervical: n=505 (48.9%); Nonambulatory noncervical: n=351 (34%); Ambulatory: n=176 (17.1%) Methodology: Participants identified from outpatient records of three hospitals were surveyed 5 times over a 20-year period in 1993, 1998, 2003, 2008, and 2013 to examine the relationship between marital status and SWB over time for people with chronic SCI. Outcome Measures: Life Situation Questionnaire-Revised (LSQ-R)	At baseline: DSW and S individuals had significantly lower scores in home satisfaction and vocational satisfaction than the MC individuals, and encountered significantly more problems in isolation, distress, environmental barriers, and money. Over 20 years… The MC group enjoyed the best SWB at baseline, but their home satisfaction and global satisfaction declined over time and their social isolation increased slightly. DSW had a greater rate of improvement; with increased home satisfaction, decreased isolation, distress, environmental barriers, money problems, and opportunities problems. For the S group, SWB trajectories were very similar to the MC individuals, except they experienced a significant increase in health problems. Each participant averaged 3 measurements (out of the 5) so attrition might have led to biases resulting from unobserved, nonrandom loss of respondents.
Cao et al. 2015; USA Level 2 Longitudinal N=1635	Population: 1635 people with SCI (74% M): Baselineà age at injury (years)= 32.98±13.83; YPI= 12.79±9.61; Nonambulatory cervical (34.8%), nonambulatory noncervical (33.95%), ambulatory (31.25%) Participants were recruited from a Southeastern United States hospital. Methodology: Subjective health and environmental barriers experienced by the participants were measured at baseline (time 1) and at a follow-up survey 5 years later (time 2). Outcome Measures: Craig Hospital Inventory of Environmental Factors- Short Form (CHIEF-SF), two health status items from the Behavioral Risk Factor Surveillance System Survey Questionnaire (physical health “not good” and mental health “not good”).	Average days physical health “not good” within the past 30 days increased from 6 days to 7 days at the 5-year follow up (P<.01). No significant changes for days mental health “not good”. 19.7% of participants reported impacts from policies barriers; 46% reported impacts from physical/structural barriers; 13.2% encountered work/school barriers; 22.4% had attitudes/support barriers; and 26% reported services/assistance barriers. After controlling for sex, race, age at injury, years since injury, and injury severity, the physical and structural barriers, and services and assistance barriers measured at baseline significantly predicted subjective physical and mental health measured at follow-up.
January et al. 2015; USA Level 2 Longitudinal N=177	Population: 177 participants with SCI (62.1% M); Mean age (y)=33.5±7.1 (range 19-50); Mean age at injury (y)=13.5±4.6 (range 0-18); Mean injury duration (y)=19.5±8.2 (range 1-43); tetraplegia (56.5%); complete injury (70.6%) Adults who sustained a SCI before 19 years of age were recruited from the Shriners Hospitals for Children SCI database. Methodology: Annual interviews were completed by participants between 2010-2014 to examine the prevalence of poor sleep quality in adults with pediatric-onset SCI and to asses the clinical correlates. Sample norms for controls and sleep-disordered patients were taken from published norms established for the PSQI. Outcome Measures: 12-item Short-Form Health Survey Version 2 (SF-12v2), Beck Anxiety Inventory (BAI), Patient Health Questionnaire (PHQ-9), Satisfaction With Life Scale (SWLS)	Sleep quality appeared to deteriorate with increased age, and was independently related to measures of psychosocial health and well-being. Problems initiating and maintaining sleep were the most frequently mentioned sleep disruptions, but difficulties sleeping due to pain and bathroom needs were also prevalent. Both increased age and tetraplegia were significantly associated with poor sleep. Activity-interfering pain and general health were also strongly associated with poor sleep. Adults with pediatric-onset SCI reported significantly more sleep difficulties across domains than the standardized sample norms, but significantly fewer problems than the sleep-disorder patient norms. Sleep difficulties were fairly common; with half of participants self-reporting poor sleep quality within the last month (51.4%).
Marcel et al. 2015; The Netherlands Level 5 Cross-sectional N=128	Population: 48 participants (64.6% M) sustaining an SCI at <50 years of age: mean age (y)=68±4, mean age at injury (y)=33.5±20, mean TSI (y)=34±18; nontraumatic SCI (33.3%); paraplegia (65.2%), incomplete injury (41.7%); 79 participants (63.3% M) sustaining an SCI at ≥50 years of age: mean age (y)=70±8, mean age at injury (y)=62±10±, mean TSI (y)=8±8; nontraumatic SCI (54.4%); paraplegia (67.1%), incomplete injury (60.8%). Methodology: Surveys were completed to study levels of participation and life satisfaction in individuals with SCI aged 65 or older and to analyze differences in participation and life satisfaction scores between individuals injured before or after 50 years of age. Outcome Measures: Frequency scale of the Utrecht Scale for Evaluation of Rehabilitations-Participation (USER-P), 5 items from the World Health Organization Quality of Life abbreviated form (WHOQOL-BREF).	Participants injured <50 years of age reported higher participation and life satisfaction scores than participants injured at older ages. The strongest difference was seen in productive activities. Participants injured before 50 years of age performed unpaid work and household work more often than participants injured later in life. Higher life satisfaction scores were positively associated with a longer time since injury, lower age at onset, higher education, having traumatic SCI, and having fewer SHCs. In the multiple regression analyses, only lower age at onset and higher education were significant independent determinants of life satisfaction, and the amount of explained variance was low. Higher participation scores were positively associated with lower current age, a longer time since injury, lower age at onset of SCI, higher education, male gender, traumatic SCI, higher functional status, and fewer SHCs. In the multiple regression analyses, lower current age, higher education, and having paraplegia were significant independent determinants of participation.
Silverman et al. 2015; USA Level 2 Longitudinal N_total=1594	Population: 1594 participants (64% F) that included people with SCI (n=414), MD (n=282), MS (n=509), and PPS (n=389); Mean age (y)=56.0±12.9 (range 20-94); mean disability duration (y)=15.3±10.5 Methodology: Community-dwelling individuals aging with a physical disability (MS, MD, PPS, or SCI) were sent an initial survey (t1) and those who completed and returned it were sent 3 more surveys over the next 3 years, approximately 1 year apart (t2, t3, t4) to investigate the effects of resilience on change in functional outcomes over time. Outcome Measures: Patient Health Questionnare-9 (PHQ-9), Patient Reported Outcomes Measurement Information System (PROMIS), Connor-Davidson Resilience Scale (CD-RISC-10)	Greater baseline resilience predicted a decrease in depressive symptoms and an increase in social functioning 3 years later but was not associated with change in physical functioning.
Cao et al. 2014; USA Level 2 Prospective Cohort N=863	Population: 863 people with SCI (72% M, 28% F) recruited from a rehabilitation hospital in Southeastern United States; Mean age (years): 38.86±11.82; Mean years post injury: 9.4 ±6.31; Ambulatory 22.75%, non-ambulatory: 77.25% Methodology: A baseline survey (1997-1998; Time 1) and a follow up survey (2008; Time 2) was completed 10 years apart to examine the relationship between unmet expectations and depressive symptoms among people with chronic SCI. 2 groups were formed: those that experienced unmet expectations (N=421), and those with met expectations (N=419) Outcome Measures: 22-item Older Adult Health and Mood Questionnaire (OAHMQ), “unmet expectations of adjustment” 10-point ladder (1=worse, 10=best adjustment), Reciprocal Social Support Scale (RSSS), probably major depression (PMD)	At baseline, there were no significant differences in OAHMQ and PMD scores between the people with or without unmet expectations. However, 10 years later, the unmet expectations groups’ OAHMQ total score was 2.2 points higher than the met expectations group, thus the unmet expectations group was more likely to have PMD at follow-up. Unmet expectations of adjustment after TSCI are positively associated with depressive
Cao et al. 2014; USA Level 2 Longitudinal N=434	Population: At baseline (1998), participants averaged 44 years of age (SD 11) and 14 years of education (SD 3); The mean (SD) age at injury was 26 (11). The majority of participants were White (81%) and male (68%) and 43% were married; 52% incomplete injuries, 48% complete injuries; Methodology: Three measurements collecting the association between household income and subjective well-being (SWB) were collected over a period of 10 years (1998, 2003, 2008) Outcome Measures: Life Situation Questionnaire-Revised (LSQ-R), including life problems/environmental barriers subscales, annual household income (less than $10K, $10K-$25K, $25K and more)	Health problems for all 3 income groups increased to some extent during 10-year period, but the low-income group appeared to increase faster than other groups. Global satisfaction mean scores for all groups dropped across 10 years. Over 10 years, participants with an income of $25K or more experienced a decrease in environmental barriers, whereas those with an income less than $10K had a modest increase. The low-income group’s vocational satisfaction increased over 10 years, whereas the rates of change for the other 2 groups did not change significantly.
Hernandez et al. 2014; USA Level 2 Longitudinal N=662	Population: 144 participants in the SCI group: mean age (y) = 37.33±15.02; 260 participants in the burns group: mean age (y) = 39.97±16.97; 260 participants in the IAF (interarticular fractures) group: mean age (y) = 44.33±17.83 Participants were a subset of a larger longitudinal study conducted by the Injury Control Research Center at the University of Alabama. Methodology: Outcome measures were administered 12, 24, 48, and 60 months post discharge to study life satisfaction trajectories over the first 5 years following medical discharge for an SCI, burn injury, or IAF. Data was collected through interviews. Outcome Measures: Functional Independence Measure (FIM), Family Satisfaction Scale (FSS), Life Satisfaction Index (LSI), marital status	There were significant FIM score differences between injury groups. At initial measurement, individuals with an SCI had significantly greater functional impairment than individuals with an IAF or burns. Over time, greater functional impairment significantly predicted lower life satisfaction, regardless of injury type. However, this association diminished when marital status and family satisfaction were entered into the models. Greater family satisfaction and being married predicted greater life satisfaction across time. Being single or separated significantly predicted lower life satisfaction, regardless of injury type.
Erosa et al. 2014; USA Level 2 Longitudinal N=144	Population: 144 people with SCI (108 M, 36 F); Mean age (y)=37; 35.4% single, 45.1% married, 11.1% divorced, 4.2% separated, 4.2% ‘other/unknown’; Methodology: A trained interviewer collected data from individuals at 12, 24, 48, and 60 months post discharge. A contextual model was tested to investigate the mediating effects of participation on the predictive relationships of functional impairment, family satisfaction, and pain to QoL following traumatic SCI. Outcome Measures: 12 months: Functional Independence Measure (FIM), Family Satisfaction Scale (FSS) 24 months: Presence of pain (yes/no) 48 months: Craig Hospital Assessment and Reporting Technique (CHART), CHART Mobility subscale, CHART Social Integration subscale, 60 months: Life Satisfaction Index (LSI), self rated health status	Greater functional impairment and pain were predictive of less participation. This relationship was mediated by mobility and social integration. Participation significantly predicted both life satisfaction and self-rated health status. Greater functional ability increased the likelihood of greater mobility, activity in social pursuits, and accompanying opportunities to socialize years later.
Krause & Bozard 2012; USA Prospective Longitudinal Level 2 N=64	Population: 64 participants with traumatic SCI (88% male, 12% female); mean (SD) age at follow-up in yrs: 61.5(7.2); mean (SD) YPI in yrs: 41.4(4.9). Methodology: Participants were enrolled in 1973 from a specialty hospital in the Midwestern United States and assessed again approximately 35 years later. Data were collected by mailed survey. Outcome Measures: The Life Situation Questionnaire.	Overall, social participation decreased over time, although the sitting tolerance and hours spent in gainful employment increased. Satisfaction with employment improved over time, whereas satisfaction with social life, sex life and health declined. Self-reported adjustment improved, but the prediction of future adjustment in 5 yrs declined.
Pershouse et al. 2012; Australia Level 3 Case-control Level 2 N=270	Population: 270 persons with traumatic SCI (81% male); mean (SD) age in yrs: 43.3(11.4), range 20-76; mean (SD) age at injury in yrs: 27.3(9.7). Methodology: Data were collected via telephone interviews and written questionnaires annually over 5 years, between 2004 and 2008, across 6 strata comprising participants grouped according to time since injury (<5y, 5–9y, 10–14y, 15–19y, 20–24y, ³25y). Outcome Measures: World Health Organization Quality of Life-8 (WHOQOL-8); Community Integration Measure (CIM)	Quality of life remained relatively stable across the lifespan Participation, as measured by the CIM, showed a significant increase with time since injury, but not when adjusted for the American Spinal Injury Association Impairment Scale, income level, and living situation.
van Leeuwen et al. 2012; Switzerland Prospective longitudinal Level 2 N=206	Population: 206 persons with recently acquired SCI; Age: range 18-65 yrs Methodology: measurements at the start of active rehabilitation, after 3 mos, at discharge, 1, 2, and 5 years after discharge. Outcome Measures: Mental Health Index (MHI-5)	Levels of mental health increased between the start of active rehabilitation and 3 mos later, remained stable thereafter, and increased again between 2 and 5 years after discharge. Latent class growth mixture modeling revealed 5 mental health trajectories: (1) high scores (above 80) at all time-points (52%), (2) low scores (£60) at all time-points (4%), (3) early recovery from 40 to scores above 70 (13%), (4) intermediate scores from 60 to scores above 70 (29%), and (5) severe deterioration of scores above 70 to scores below 30 (2%).
DeVivo & Chen 2011; USA Longitudinal Level 2 N=1591	Population: 1591 individuals with paraplegia and tetraplegia, complete and incomplete; Mean age 39.5 yrs; Mean YPI 1.0. Methodology: Participants studied at seven time points at 1, 5, 10, 15, 20, 25, and 30 YPI. Outcome Measures: Satisfaction with Life Scale.	Life satisfaction consistently increased over the 30 years of observation.
Kalpakjian et al. 2011; USA Longitudinal Level 2 N=6141	Population: 6141 individuals with paraplegia and tetraplegia, complete and incomplete; Mean age 39.0 yrs; Mean YPI 1.0; Male = 4864. Methodology: Participants studied at 1 to >5 time points at 1 and 5 YPI, and every 5 years after. Outcome Measures: Satisfaction with Life Scale.	Life satisfaction was shown to improve over time. Life satisfaction of those separated or divorced increased over time and did so to a greater degree for women.
Van Leeuwen et al. 2011; The Netherlands Longitudinal Level 2 N=206	Population: 206 participants with SCI (153M 53F); Mean age 41.5 yrs, range 26.2-56.9. Methodology: Participants studied at six points in time (Beginning of rehabilitation; 3 months after beginning of rehabilitation; at discharge, 1 year post discharge, 2 years post discharge and 5 years post discharge). Outcome Measures: Two study specific questions: 1) current life satisfaction; and 2) life satisfaction now vs prior to SCI.	56 participants (27%) had consistently low, yet improving, life satisfaction. 34 participants (17%) had consistently high life satisfaction. 63 participants (31%) had intermediate scores. 48 participants (23%) improved their life satisfaction. 5 participants (2%) had deteriorating life satisfaction.
Mitchell & Adkins 2010; USA Longitudinal with AB controls Level 2 N SCI=26 N controls=38	Population: 26 individuals with SCI, mean(SD) age at baseline 50.3(11.2) yrs, mean(SD) YPI at baseline 25.2(14.2); YPI >5 yrs, no cognitive impairment. 38 age matched AB controls, mean(SD) age at baseline 50.4(11.5) yrs. Methodology: Data collection at two time periods 5 years apart from 1999 – 2002, and 2004 -2007. Outcome Measures: Potential predictors of Self-related health (SRH) included: aging, self-report Activities of Daily Living (ADL) and independent activities of daily living (IADL); 4-point pain severity scale; Fatigue Severity Subscale from the Fatigue Assessment Instrument; and the Older Adult Health and Mood Questionnaire.	SCI participants and control group declined significantly in SRH overtime. SRH ratings at follow-up significantly differed by group: decrease in excellent/very good category from 57.7% to 24.6% for the SCI group and decrease from 76.3% to 63.2% in control group. Aging has greater influence on SRH in people with SCI than on those without a SCI. Over time, the SCI group differed significantly from comparison group on all health measures: increase in health conditions, increase in fatigue, decrease in ADLs and decrease in IADLs.
Mortenson et al. 2010; Canada Longitudinal Level 2 N=93	Population: 93 participants with SCI (83M 10F); Mean age 39.6 yrs, range 18-78 yrs; Mean YPI >0.25 post discharge. Methodology: Participants studied at two time points between 1999 and 2003 (3 months and 15 months post discharge). Outcome Measures: Quality of Life Index.	No QoL differences were observed between time points. Average QLI scores were 19.66 out of 30.
Savic et al. 2010; Britain Longitudinal Level 2 N=122	Population: 122 participants with SCI (103M 19F); Mean age 48.0 yrs; Mean YPI 26.8. Methodology: Participants studied at six points in time (1990; 1993; 1996; 1999; 2002; and 2006). Outcome Measures: Life Satisfaction Index.	76% consistently rated good or excellent QoL over 16 years. Differences in life satisfaction were observed over the 16 yr period. Life satisfaction was highest in the final (2006) follow up.
Krause et al. 2009; USA Longitudinal Level 2 N=250	Population: 250 individuals (142M 108F) with SCI (Tetraplegia = 133; Paraplegia = 117); Mean age = 42.5 yrs; YPI = 14.0. Methodology: Participants studied at two points in time 6 years apart. Outcome Measures: Life Satisfaction Questionnaire.	Differences in the satisfaction domain did not systematically or significantly increase or decrease over six years.
Van Koppenhagen et al. 2009; The Netherlands Longitudinal Level 2 N=222	Population: 222 individuals with SCI (Tetraplegia complete= 54, Paraplegia complete= 96, Other = 38); Mean age (range) = 41.5 yrs (18-65); YPI = <0.9; Male = 165. Methodology: Participants studied at four points in time (Beginning of rehabilitation; 3 months after beginning of rehabilitation; at discharge and 1 yr post discharge). Outcome Measures: Two study specific questions: 1) current life satisfaction; and 2) life satisfaction now vs. prior to SCI.	Individuals unsatisfied with their life decreased from 74.6% at time 1 to 49.3% at time 4. Individuals with deterioration in life satisfaction after SCI decreased from 85.8% at time 1 to 75.3% at time 4. There was a significant increase in life satisfaction total scores between times 1 and 3, and stable scores from times 3 to 4.
Krause & Coker 2006; USA Longitudinal Level 2 N initial=256 N final=78	Population: 78 respondents with SCI, mean(SD) age 55.7(7.6) yrs and mean(SD) YPI 35.8(4.7) at follow-up; mean(SD) age at enrollment 35.1(13.9) yrs and mean(SD) YPI at enrollment 9.7(6.9). Methodology: Participants assessed 3 times over a 30-year period (1973, 1988, 2002). Outcome Measures: Life Situation Questionnaire (LSQ).	During the first 15-year period, an increase was seen in educational and employment outcomes, satisfaction with employment, and adjustment. During the last 15 years, satisfaction with social life and sex life decreased.
Bushnik & Charlifue 2005; USA Longitudinal Level 2 N=63	Population: 63 individuals, C1-C4, sustained at a mean(SD) age of 24.2(2.2) yrs; Mean(SD) age (yrs): phase 1 = 29.7(8.1), phase 2 = 37.4(8.4), phase 3 = 43.2(8.4); Mean(SD) YPI: phase 1 = 5.3(2.4), phase 2 = 13.1(2.5), phase 3 = 18.9(2.6). Methodology: Participants assessed 14-24 YPI over three phases of data collection. Phase 1: 1984-1985; Phase 2: 1992-1994; and Phase 3: 1997-1999. Outcome Measures: Demographics, respiratory support, assistance with care, psychosocial data, and time utilization data.	Over the three phases of data collection, most indicators of quality of life and community participation remained at high levels with little variability. For those indicators that changed, SCI itself and not necessarily a high level of injury, and/or the aging process appeared to account for the change over time. Economics and advances in technology were associated with changes in other indicators.
Krause & Broderick 2005; USA Longitudinal Level 2 N initial=161 N final=95	Population: 95 individuals with SCI, mean(SD) age 53.8(9.2) yrs and mean(SD) YPI 32.2(5.6) at follow-up. Methodology: Participants assessed twice over a period of 25 years (1974 and 1998). Outcome Measures: Life Satisfaction Questionnaire (LSQ).	Employment outcomes improved; rate and satisfaction increased. Participants continue to adapt to aging with SCI (increasing adjustment scores), yet experience declines in health (increase in number of physician visits) and social activities (decrease in number of outings and visitors) that were associated with diminished satisfaction with particular areas of life (i.e. social and sex lives).
Charlifue & Gerhart 2004a; USA Longitudinal Level 2 N=178	Population: 178 individuals with SCI, age 43-83 yrs, YPI 29-55 at final phase of study. Methodology: Participants evaluated at 3 yr intervals from 1990-1999. Outcome Measures: Craig Handicap Assessment and Reporting Technique (CHART), Life Satisfaction Index (LSI-Z), Index of Psychological Well-Being (IPWB), Center for Epidemiological Studies Depression Scale (CES-D), Perceived Stress Scale (PSS).	Community reintegration decreased over time – a significant relationship between perceived life satisfaction and community integration was revealed.
Charlifue & Gerhart 2004b; USA Longitudinal Level 2 N=189	Population: 189 participants with SCI; age (range) 43-83 yrs and YPI 29-55 at final phase of study. Methodology: Participants evaluated at 3 yr intervals from 1990-1999. Outcome Measures: Life Satisfaction Index (LSI-Z), Index of Psychological Well-Being (IPWB), Center for Epidemiological Studies Depression Scale (CES-D), Perceived Stress Scale (PSS); Perceived Quality of Life.	LSI-Z scores were significantly correlated with perceived quality of life 3 yrs later at all three data points. Previous scores on perceived QoL were correlated with perceived well-being 3 yrs later at all three data points. Being male, older, longer injured, and having functionally complete tetraplegia were related to worse outcomes.
Bushnik 2002; USA Longitudinal Level 2 N=58	Population: 58 individuals with complete high tetraplegia (C1-C4); Mean(SD) age (yrs): phase 1 = 29.5(8.4), phase 2 = 37.3(8.7), phase 3 = 43.0(8.7); Mean(SD) YPI: phase 1 = 4.8(2.2), phase 2 = 12.7(2.4), phase 3 = 18.5(2.4). Methodology: Participants assessed up to 24 YPI over three phases of data collection. Phase 1: 1984-1985; Phase 2: 1992-1994; Phase 3: 1997-1999. Outcome Measures: High Tetraplegia Questionnaire (HQQ) consisting of questions regarding assistance with care, respiratory support, special equipment, and costs, participation in work and leisure activities; Modified Level of Free Time Activities Scale; Rosenberg Self-Esteem Scale (RSE).	An association between access to modified vans and perceived QoL was revealed. The proportion of individuals who rated their lives as “good or excellent” was higher in all three phases for those with modified vans, reaching statistical significance in Phases 2 and 3. No significant changes in equipment utilization nor self-esteem or subjective QoL across time periods. Participation in at-home leisure activities (e.g. reading, listening to the radio) significantly decreased from Phase 1 to Phase 2. Older individuals spent fewer hours per day out of bed and fewer days per week out of the house than younger participants in Phase 1. In Phase 3, older individuals participated in fewer leisure activities at home than younger participants. Income increased over time for the sample.
Putzke et al. 2002b; USA Longitudinal Level 2 N=270	Population: 270 participants with SCI. Methodology: Participants studied at 1 and 2 years post-injury. Outcome Measures: Satisfaction with Life Scale (SWLS), Craig Handicap Assessment Reporting Technique (CHART), Short-Form 12 (SF-12).	Those developing pain (i.e. pain-free at year 1 and in pain at year 2) reported decreased life satisfaction, physical and mental health, whereas those resolving pain from year 1 to year 2 reported an increase in the same domains. Change in pain interference status was unrelated to change in self-reported handicap.
Charlifue et al. 1998; USA Longitudinal Level 2 N=227	Population: 227 individuals with SCI, YPI at enrollment >20, age (range) at injury 15-55 yrs, mean age 54.9 yrs and mean YPI 30.0 at final phase of study. Methodology: Participants contacted twice over a 3-year period (1990 and1993). Outcome Measures: Health and functional status (physical exams), self-perceived health, Functional Independence Measures (FIM), Craig Handicap Assessment and Reporting Technique (CHART), Life Satisfaction Index (LSI-Z), Index of Psychological Well-Being (IPWB), the Center for Epidemiological Studies Depression Scale (CES-D), Perceived Stress Scale (PSS), Current Problem Questionnaire.	LSI-Z scores decreased over time. These were statistically significant only for the older age groups, those injured less than 30 yrs or more than 40 yrs, and those with complete paraplegia. Those who scored high on the LSI-Z experienced higher social integration 3 yrs later. Those who drank more also had higher social integration scores. Those who reported feeling generally healthy were only ¼ as likely to experience fatigue 3 yrs later, while those who reported fatigue previously were more than 5 times more likely to still feel fatigue 3 yrs later.
Krause 1998; USA Longitudinal Level 2 N=114	Population: 114 individuals (95M 19F) with SCI (Tetraplegia = 67, Paraplegia = 47); Mean age = 29.9 yrs; Mean YPI = 8.7. Methodology: Participants studied at four points in time (1974; 1985; 1989; 1994). Outcome Measures: Life Situation Questionnaire (LSQ).	Satisfaction with employment increased over 20 yrs. Satisfaction with living arrangements improved at each time point, and general health worsened (n.s.). Satisfaction with finances, social and sex lives worsened overall (n.s.).
Krause 1997; USA Longitudinal Level 2 N initial=347 N final=235	Population: 235 participants with SCI, mean(SD) age at injury 23.1(9.1) yrs, mean(SD) age and YPI at last data collection point: 46.7(10.7) yrs and 23.6(7.4). Methodology: Participants assessed twice over a 9-year period (1985, 1994). Outcome Measures: Life Situation Questionnaire (LSQ).	Objective aspects of life situation remained stable (i.e. frequency of visitors and social outings, sitting tolerance) or increased (i.e. employment rate, hours worked) over the study period; however, subjective wellbeing diminished. Participants were less satisfied with their lives, reported more problems with dependency, and were more pessimistic when predicting their future adjustment.
Stensman 1994; Sweden Longitudinal Level 2 N=17	Population: 17 participants with SCI (15M 2F); Mean age = 32.4 yrs; Mean YPI = 0.5. Methodology: Participants studied at six time points at 0.5, 1, 2, 3, 4, 5 YPI. Outcome Measures: Single item to measure perceived QoL.	5 participants (29%) had consistently high QoL. 6 participants (35%) had low QoL between 0.5 and 3 yrs, and improvements thereafter. 2 participants (12%) had variable QoL over time, influence by variable pain. 4 (24%) had consistently low QoL, influenced by constant pain.
Krause 1992; USA Longitudinal Level 2 N=135	Population: 135 individuals (110M 25F) with SCI (Tetraplegia = 91; Paraplegia = 44); Mean age = 30.6 yrs; Mean YPI = 9.2. Methodology: Participants studied at two points in time (1974; 1989). Outcome Measures: Life Situation Questionnaire (LSQ).	Satisfaction with employment and finances increased over 15 years. Satisfaction with social life and general health stayed the same.
Crewe & Krause 1990; USA Longitudinal Level 2 N=154	Population: 154 individuals (124M 30F) with SCI (Tetraplegia = 95; Paraplegia = 58); Mean age = 32.0; YPI = 10.0. Methodology: Participants studied at two points in time points (1974; 1985). Outcome Measures: Life Situation Questionnaire.	Satisfaction with employment increased significantly over 11 yrs. Satisfaction with living arrangements, finances, and social life did not change over time. Satisfaction with sex life and satisfaction with general health remained the same.
Lundstrom et al. 2013; Sweden Level 5 Cross-sectional N=97	Population: N=97 (56.7 % M), mean age (y)= 43.5±12 (range 19-69); mean TSI (y)=7.3±5.6. Age was divided into: group 1 (age 19-44); group 2 (age 45-69). TSI was divided into: group 1: 2-4 years; group 2: 5-8 years; group 3: 9-31 years. Methodology: Participants were given a survey to explore and describe the leisure activities of people with tSCI and their relationship to interest, performance, and well-being. Outcome Measures: Two-part postal survey that investigated sociodemographic variables and injury characteristics, and an interest checklist with 20 areas of leisure activites.	-44) were least likely to engage in ball sports and equipment sports for their own interest and well-being
Kemp & Krause 1999; USA Cross-sectional Level 5 N SCI=177 N controls=62	Population: 177 participants with SCI; mean age at onset 25.5 yrs; mean age at enrollment 40.1 yrs. 62 healthy participants; mean age at enrollment 63.7 yrs. Methodology: Comparison of life satisfaction and depression between the 2 groups. Outcome Measures: 10-item scale to evaluate life satisfaction.	The AB comparison group had significantly greater satisfaction than the SCI group in life areas related to: health, health care, finances, friendship, emotional health, housing, work or time use, leisure, and overall life.
Barker et al. 2009; Australia Cross-sectional Level 5 N=270	Population: 270 individuals (220M 50F) with SCI (Tetraplegia = 107; Paraplegia = 100; Others=63); Mean age (range) = 43.0 (20-76); YPI (range) = 15.0 (9-21). Methodology: Comparison of QoL between SCI and AB groups, and group differences were observed by age. Outcome Measures: World Health Organization Quality of Life Assessment Instrument (WHOQOL)-Brief.	QoL was lower than the Australian norm. Overall, mean scores in each of the 4 domains (Physical health, Psychological health, Social Relationships, Environment) were lower for individuals with SCI. Analyses by age group showed significant differences between SCI and norm data in all domains, with the exception of the 60s and over age group that differed in the Physical health domain only. No differences were detected between people with SCI of different ages and times since injury.

Discussion

Aging is a complex process that does not only encompass biology. Environmental factors also change over time, which may be particularly important to persons with SCI, because they not only face physical limitations associated with their SCI, but also injury-related social and economic changes (Krause & Coker 2006). For example, in a series of papers reporting on the same cohort at different time points over a period of 30 years, there were significant improvements with satisfaction with employment and finances over time (Crewe & Krause 1990; Krause 1992; Krause 1998; Krause & Broderick 2005; Krause & Coker 2006), whereas satisfaction with both social and sexual relationships decreased (Krause 1997; Krause & Broderick 2005; Krause & Coker 2006). Similarly, Bushnik & Charlifue (2005) observed changes related to economics and technology, but not related to SCI or aging per se. For example, letter writing, which probably included emails, increased in the sample over time because home computing had likely become more common. Although not significant, the high percentage of persons who switched to a portable ventilator or pneumobelt from a fixed ventilator may have improved community reintegration for these individuals. As well, the finding that economic self-sufficiency steadily improved with time (e.g. Charlifue & Gerhart 2004a; Krause & Broderick 2005; Krause & Coker 2006) supports Bushnik’s (2002) speculation that increased economic standing may improve community reintegration. In the case of Bushnik’s (2002) sample, improved financial status enabled better access to adaptive equipment (e.g. modified vans).

Conversely, level of community reintegration for Charlifue & Gerhart’s (2004a) sample did not significantly change over time, but this may have been due to sample differences between the studies and that the time between data collection intervals in the other studies reviewed were further apart. As well, the individuals in Charlifue and Gerhart’s (2004a) study were at least 20 years post-injury when they entered the study. At 20 years post-injury, it is likely that routines and strategies for community participation have been well-established, and are not likely to dramatically change over 3 year periods. However, an understanding of environmental factors is important for assessing QoL since there is evidence that an individual’s adjustment over time is influenced by corresponding environmental changes (Krause & Sternberg 1997).

With regards to change in activity patterns, Bushnik and Charlifue (2005) attributed the changes to the natural progression of time utilization from external social activities associated with youth (e.g. card games with friends) to other activities (e.g. spending time with family). Further, the reported declines in activity by the SCI cohorts as they aged (e.g., Bushnik 2002, Charlifue and Gerhart 2004a, and Krause & Broderick 2005) might be similar to declines in activity patterns in the general population (Christensen et al. 1996; Bukov et al. 2002).

One of the main strengths of the studies by Krause (1997), Krause and Broderick (2005), and Krause and Coker (2006) is they assessed whether there were any differences between their current sample and those who were lost to follow-up. Based on these analyses, clear survivor effects emerged in both studies as the characteristics of respondents (persons who participated in both data collection periods) at Time 1 were younger, younger at age of SCI-onset, were less years post-injury, had higher levels of education, more likely to have cervical injuries, greater sitting tolerance, and had more social outings than non-respondents (persons who only participated in the first data collection period). These findings highlight that some care should be taken when interpreting the findings from these studies as it may only reflect survivors, and those who continued to participate.

The findings appear to provide some mixed evidence regarding the stability of QoL/life satisfaction over time. In some cases QoL/life satisfaction remained stable (i.e., Charlifue et al. 1998; Charlifue et al. 1999; Charlifue & Gerhart 2004b; Savic et al. 2010; Pershouse et al. 2012), or decreased over time (i.e., Krause 1997; Charlifue et al. 1998). Similarly, Mitchell & Adkins 2010 that aging has greater negative influence on self-rated health in people with SCI than on those without a SCI over time. In other studies QOL/ life satisfaction improved with over (Stensman 1994; Kemp & Krause 1999; Bushnik 2002; Putzke et al. 2002b; Bushnik & Charlifue 2005; Krause & Coker 2006; van Koppenhagen et al. 2009; DeVivo & Chen 2011; Kalpakjian et al. 2011; van Leeuwen et al. 2011). Likewise, mental health has also been reported to improve longitudinally (van Leeuwen et al. 2012).

The discrepancies in these studies may potentially be attributed to theoretical and methodological differences. For instance, the study by Charlifue et al. (1998) was the only study that explicitly provided a theoretical model for assessing life satisfaction. Specifically, Charlifue and colleagues (1998) framed aging with SCI within a global thesis of function, which took into account physical, psychological, and environmental factors. Several studies with lower levels of evidence predicting life satisfaction have used other models that incorporate a variety of domains thought to impact on QoL (i.e., Pierce et al. 1999; Richards et al. 1999; Tonack et al. 2008). Unfortunately, Charlifue et al. (1998) did not provide a clear rationale for including specific predictor variables in their models. A larger theoretical concern is the issue of response shift (also known as recalibration, reprioritization, and reconceptualization; Schwartz & Spangers 2000), which refers to a dynamic process where an individual undergoes simultaneous changes in their internal standards, values, and conceptualizations of QoL in response to health and physical functioning changes (Tate et al. 2002). Ambiguous or paradoxical findings can occur because of differences among people or changes within people regarding internal standards, values, or conceptualization of health-related QoL (Schwartz et al. 2007). As a result, the psychometric properties (e.g., validity and reliability) of measurement tools can be affected (Schwartz et al. 2007).

In terms of methodological differences, because the samples in each of the studies had different mean ages and YPI it is not surprising that there are discrepancies in reported QoL. However, when examining the QoL results by an aging parameter, YPI for example, a common finding was that regardless of age, individuals with relatively new SCI (i.e.≤5 YPI) are more likely to experience improvements to their QoL (Stensman 1994; Kemp & Krause 1999; Bushnik 2002; Putzke et al. 2002b; Bushnik & Charlifue 2005; Krause & Coker 2006; van Koppenhagen et al. 2009; DeVivo & Chen 2011; Kalpakjian et al. 2011; van Leeuwen et al. 2011) than individuals with longer term SCI (i.e., ≥6 YPI) who consistently report high and stable QoL levels (i.e., Charlifue et al. 1998; Charlifue et al. 1999; Charlifue & Gerhart 2004b; Savic et al. 2010). That is, after sustaining a traumatic SCI, the QoL of these individuals may be low and have more room to improve than those individuals with longer term SCI. In fact, Dijkers (2005) noted that the wellbeing after SCI reaches a plateau at the end of the adjustment period, which is estimated to last from two to five years (Dijkers 2005). Similarly, Whalley-Hammell (2007) reported that after a four year adjustment period, individuals with SCI feel as though as they live a normal life, and have the same problems as everyone else (Whalley-Hammell 2007). In this review, there one study however that observed no changes in QoL among individuals with ≤5 YPI (Mortenson et al. 2010). Mortenson et al. (2010) argued that the individuals may have already adjusted and experienced a response shift prior to the baseline assessment.

Although age of SCI onset does not appear to limit the potential high QoL, there are likely age-related factors that may potentially influence QoL. For example, in studies with samples with mean ages in the 20s, individuals were found to have greater improvements in life satisfaction and QoL if they were students, lived independently, had a lower level injury, had overcome past medical problems, and if they had accessible vans for transportation (Barker et al. 2009; Sakakibara et al. 2012). Among individuals in their 30s, both Putzke et al. (2002b) and Stensman (1994) found QoL to be influenced by amount of pain and interference with pain (Putzke et al. 2002b; Stensman 1994),and Kalpakjian et al. (2011) found the relationship between life satisfaction and YPI to vary depending on marital status and gender (Kalpakjian et al. 2011).

Furthermore, the nature of the control group can lead to different interpretations of the results. A strength of Kemp and Krause’s (1999) was the use of an able-bodied, and a control group with disability (i.e., polio) when examining issues of QoL after SCI as it provides some context to the extent of some problems for persons post-SCI (i.e. levels of depression). However, the characteristics of the control groups were significantly different to the group with SCI on some key factors. For instance, the able-bodied and polio groups were significantly older (p< 0.01) and had higher levels of education than the group with SCI (p< 0.05). As well, the polio group was comprised mostly of females, had a mean pediatric age of onset, was 50.9 years post-polio, and 90% were Caucasian, whereas the SCI group was comprised of mostly males from culturally diverse backgrounds, and who had an adult age of onset, and were only 14.5 years post-injury. This limitation was addressed in the study, but highlights that the findings should be interpreted with caution since many socio-demographic and historical factors may have influenced levels of depression and life satisfaction. Nonetheless, the finding that persons with SCI have lower QoL compared to the able-bodied population is consistent with other studies that did not meet the SCIRE inclusion criteria (Kemp & Ettelson 2001).

Finally, although a couple of studies reported declines in QoL over time (Krause 1997; Charlifue et al. 1998), subsequent papers focusing on the same cohorts at longer lengths of follow-up reported different results. For example, Charlifue et al. (1998) first reported that after 3 years of observation 76% of the sample consistently rated their overall QoL as either good or excellent, but that there were significant decreases in life satisfaction, as measured by the life satisfaction index (LSI), among older individuals, those with <30 YPI and >40 YPI, and those with complete paraplegia (Charlifue et al. 1998). At a follow up thirteen years later, Savic et al. (2010) similarly reported that 76% of the sample consistently reported overall QoL as good or excellent, with the highest life satisfaction reported at the last time point (Savic et al. 2010). Similarly, over two time points 9 years apart, Krause (1997) reported diminished satisfaction related to social and sex lives, as measured by the life situation questionnaire (LSQ)* (Krause 1997). Lower satisfaction is corroborated in papers by Krause and Broderick (2005) and Krause and Coker (2006) which used observations from the same cohort at different lengths of follow-up (Krause & Broderick 2005; Krause & Coker 2006). However, these two papers in addition to Crewe and Krause (1990), Krause (1992), and Krause (1998), all reported significant increases in satisfaction related to employment among the same cohort over various lengths of time (Crewe & Krause 1990; Krause 1992; Krause 1998). In general, the overall and common finding from studies that followed the same cohorts over time is that global QoL tends to remain high and stable over time but when considering specific areas of QoL, fluctuations exist with some domains increasing in importance (e.g. employment) and other decreasing (e.g. social and sex lives) (Krause & Bozard 2012).

*Note: Krause 1997 used a modified version of the LSQ. Using this version, the authors also observed significant declines in satisfaction related to family relationships, emotional adjustment and control over life.

Conclusion

There is level 2 evidence from one cohort study (Mitchell & Adkins 2010) that aging has greater influence on self-rated health in people with SCI than on those without a SCI.

There is Level 4 evidence from four longitudinal studies (Bushnik 2002; Bushnik & Charlifue 2005; Krause & Broderick 2005; Krause & Coker 2006) that changes in environmental factors over time (i.e., economics, technology) may influence QoL in persons with SCI rather than the aging process per se.

There is Level 4 evidence from three longitudinal studies (Charlifue & Gerhart 2004a; Bushnik & Charlifue 2005; Krause & Bozard 2012) that community reintegration and social participation declines with age after SCI. However, these changes in community reintegration may be similar as compared to the aging general population.

There is Level 4 evidence from seven longitudinal studies (Crewe & Krause 1990; Krause 1992; Krause 1997; Krause 1998; Krause & Broderick 2005; Krause & Coker 2006; Krause & Bozard 2012) that selected domains of life satisfaction change (i.e., social life, sex life, and health decrease, and employment, finances, and adjustment increase) as one ages with an SCI. It may be that these changes in satisfaction of certain domains are comparable to changes in the general population.

There is Level 5 evidence from one cross-sectional study (Kemp & Krause 1999) that age of SCI-onset may be an influential factor on life satisfaction.

There is Level 4 evidence from one longitudinal study (Charlifue & Gerhart 2004b) that previous perceptions of life satisfaction are predictive of later perceptions of life satisfaction.

There is Level 5 evidence from two cross-sectional studies (Kemp & Krause 1999; Barker et al. 2009) that life satisfaction is lower for persons with SCI compared to the general population.

There is Level 4 evidence from two longitudinal studies (Stensman 1994; Putzke et al. 2002a) that previous reports of pain interference after SCI, irrespective of age, are predictive of later pain interference.

There is level 4 evidence from 10 longitudinal studies that individuals with ≤5 YPI have the potential to improve their QoL (Stensman 1994; Kemp & Krause 1999; Bushnik 2002; Putzke et al. 2002b; Bushnik & Charlifue 2005; Krause & Coker 2006; van Koppenhagen et al. 2009; DeVivo & Chen 2011; Kalpakjian et al. 2011; van Leeuwen et al. 2011).

There is level 4 evidence from 4 longitudinal studies that individuals with longer term SCI (i.e., ≥6 YPI) consistently report high and stable QoL levels (Charlifue et al. 1998; Charlifue & Gerhart 2004b; Savic et al. 2010). Similarly, there is Level 4 evidence from one longitudinal study (Pershouse et al. 2012) that QoL remains stable across the lifespan even in those with long-duration SCI.

Selected domains of life satisfaction (i.e. social life and sex life) may decline as one ages with a SCI. Other domains (i.e., employment and finances) may improve as one ages with a SCI. It may be that these changes in satisfaction of certain domains are comparable to changes in the general population.

Changes in environmental factors over time (i.e. economics; technology) may influence QoL in persons with SCI rather than the aging process per se.

Community participation may decline with age after SCI. However, these changes in community participation may be similar to the aging general population.

Individuals with new SCI (i.e. ≤ 5YPI) consistently report improvements to their QoL, whereas individuals with longer term SCI consistently report high and stable QoL over time.

Age of SCI-onset may be an influential factor on life satisfaction.

Previous perceptions of life satisfaction may be predictive of later perceptions of life satisfaction.

Aging has greater influence on self-rated health in people with SCI than those without a SCI.