Abstract
Context/Objective
Examine demographics, injury characteristics, objective measures of social isolation and health factors that are associated with perceived social isolation (PSI) among Veterans with spinal cord injury and disorders (SCI/D).
Design
Cross-sectional survey.
Setting
The Veterans Health Administrations (VHA) SCI/D system of care.
Participants
Veterans with SCI/D who have used the VHA health care system.
Intervention
Not applicable.
Outcomes Measures
We assessed unadjusted associations of high PSI (above population mean) vs low (normative/below population mean), and multivariable logistic regression for independent associations with PSI.
Results
Out of 1942 Veterans with SCI/D, 421 completed the survey (22% response rate). Over half (56%) had PSI mean scores higher than the general population. Among the objective measures, having a smaller social network size was associated with increased odds of high PSI (OR 3.59, P < .0001); additionally, for health factors, having depression (OR 3.98, P < 0.0001), anxiety (OR 2.29, P = 0.009), and post-traumatic stress (OR 2.56, P = 0.003) in the previous 6 months, and having 4 or more chronically occurring secondary conditions (OR 1.78, P = 0.045) was associated with increased odds of high PSI. The most commonly identified contributors to feelings of PSI included mobility concerns (63%), having a SCI/D (61%), and concerns about being a burden on others (57%).
Conclusions
Factors such as social network size may be used to identify individuals with SCI/D at risk for PSI. Additionally, by identifying mental health problems, presence of multiple chronically occurring secondary conditions, and Veteran-identified contributors of PSI, we can target these factors in a patient-centered interventions to identify and reduce PSI.
Keywords: Spinal cord injury and disorder, Perceived social isolation, Loneliness, Veteran
Introduction
The quantity and quality of social relationships play a key role in mental, behavioral, and physical health, and inadequacies in these social relationships can affect overall well-being (1–6). Throughout the literature, the construct of isolation is separated into a subjective and objective component. Perceived social isolation (PSI), often referred to as loneliness, is typically defined as a subjective measure of feeling isolated (1). Whereas objective social isolation, often referred to as social isolation, is the quantity of social interactions and relationships and is often measured by the quantity and quality of social contact and network size (3, 7, 8). Although these definitions have not achieved wide-scale consensus (9), it is generally agreed that loneliness and social isolation are independent constructs, due to their often non-significant correlation (10).
PSI is a common and growing issue with 46% of United States (U.S.) adults reporting feeling alone or isolated (11). Individuals with any mental or physical disabilities have been shown to experience significantly higher rates of PSI compared to people without a disability (12). One population that may be at risk is individuals with spinal cord injuries and disorders (SCI/D). In surveys of individuals with SCI/D, up to 55% (13) reported loneliness. People living with SCI/D may have greater risk of experiencing PSI due to varying degrees of disability attributed to comorbidities or SCI/D secondary health conditions (e.g. neurogenic bladder and bowel, urinary tract infections) (14) and environmental barriers (e.g. lack of accessibility of surroundings) (15). Additionally, 57% of Veterans reported some of the time or often feeling lonely (16). Approximately 302,000 individuals in the U.S. are living with traumatic SCI/D (17) and over 15% of traumatic and nontraumatic individuals are receiving care from the Veterans Health Administration (VHA) (18), the largest integrated healthcare system in the United States, this allows VHA a unique opportunity to provide research-based knowledge, such as factors associated with PSI, to the Veteran SCI/D community.
Feelings of PSI can influence a person’s biological pathways by heightening inflammatory response, which in turn can increase the risk of dementia (19), sleep issues (20), depression, and other cognitive, mental, and physical health outcomes (21, 22). In 2022 the American Heart Association published a statement indicating PSI is associated with a 30% increased risk of heart attacks and a 32% increased risk of stroke among the general public (23). PSI has even been shown to increase the risk of premature mortality, with a hazard ratio up to 1.22 (95% confidence interval (CI): 1.10–1.35) (24). Due to the widespread and profound threat on individuals’ health and well-being, in 2023 the U.S. Surgeon General put an advisory on loneliness and social isolation (25).
Given the association of PSI and an individual’s health and wellbeing, especially those with SCI/D, it is important to identify factors associated with PSI. Identifying associated factors with PSI can help healthcare professionals and other stakeholders identify those at risk for PSI and understand the self-reported contributors of PSI; furthermore, it can highlight areas that can be targeted for interventions to help people manage PSI, which aligns with the call to action the U.S. Surgeon General indicated in his advisory on the epidemic of loneliness and social isolation in the U.S. The aim of this study was to examine factors that are independently associated with PSI, considering demographics, injury characteristics, health factors, and objective measures of social isolation among Veterans with SCI/D.
Methods
Design/Setting/Sample. A cross-sectional, national survey was conducted with a sample of Veterans with SCI/D between October 2022 and January 2023. The sampling frame included 16,672 Veterans with SCI/D who had a recent encounter within the VHA healthcare system, and a valid residential mailing address documented. Our goal was to invite 2,000 individuals to participate, which is 12% of the sampling pool. Our power calculation determined that 376 Veterans with SCI/D were needed with a 5% margin of error at 95% confidence level. A response rate of 20–25% was assumed in the sample size calculation (26, 27), and the population of eligible Veterans with SCI/D was 16,672. The sample was identified using VHA medical record databases, including the Corporate Data Warehouse (CDW), a national repository for VHA clinical and administrative data, and the SCI/D registry, a national data registry for Veterans with SCI/D who received care within the VHA SCI/D system of care. The sample was comprised of individuals with both traumatic and non-traumatic SCI/D and could include those diagnosed with nonmalignant neoplasms resulting in neurologic deficit; vascular insults of a thromboembolic, hemorrhagic, or ischemic nature; cauda equina syndrome producing neurologic deficit; inflammatory disease of the spine, spinal cord or cauda equina resulting in non-progressive neurologic deficit; and demyelinating disease of the spinal cord. The Institutional Review Board at Edward Hines, Jr. VA Hospital approved this study and approved waived written informed consent for participants.
Eligible Veterans were mailed an introductory letter providing a brief overview of the study and invitation to participate in the survey, copy of the survey, and a postage-paid envelope. To increase response rate, we conducted a follow-up mailing approximately 6 weeks after the first mailing with non-respondents.
Survey Instrument and Variables. The survey collected data on the dependent variable, PSI, which was measured using the validated Patient-Reported Outcomes Measurement Information System short form for social isolation (PROMIS-SI) (28). The survey also collected demographics, objective social isolation measures, and health factors. A stand-alone question asked individuals to identify what contributes to their feelings of PSI.
PSI. We defined PSI as subjectively feeling left out, avoided, and detached from others and isolated even when not alone. We measured PSI using the PROMIS-SI which consists of 8 questions on a 5-point Likert scale coded from 1 (never) to 5 (always). Raw scores were converted to T-scores, with higher scores indicating greater PSI. To determine above average PSI, a score of 50, which is the U.S. general population average PSI score, was used (26). PSI was dichotomized into scores above the population mean (>50) indicating high PSI and at/below the normative population mean (< = 50) indicating low PSI.
Demographics characteristics included sex (male/female), race (white/black/other), ethnicity (Hispanic or Latino/not Hispanic or Latino), marital status (married/never married, separated, or widowed), age, and injury characteristics (including injury level, completeness, etiology, duration, and age at onset). We supplemented missing survey data on demographics and injury characteristics with data from the CDW.
Based on previous surveys conducted by the authors, objective social isolation included self-reported presence/absence of an informal caregiver (question asked if they have an informal caregiver that helps with their care) and current living arrangement (lives alone/does not live alone) (26, 27). We also asked about social network size (response options: 1 person, 2–3 people, and 4 or more, responses were dichotomized to <4 or 4+ people); modeled from Guilcher et al. (2). Frequency of meaningful contacts (question asked how often they can open up to someone to talk about their worries; response options: never/hardly ever vs. some of the time/often), and frequency of reliable contacts (question asked how often they can rely on someone to help if they have a problem; response options: never/hardly ever vs. some of the time/often) were modeled on items from the psychosocial core module on the Health and Retirement study surveys (29).
Health factors included the presence of health complications in the last 6 months (depression, anxiety, dysfunctional sleep, and post-traumatic stress). We also collected presence of SCI/D secondary health conditions (chronic pain, joint/muscle pain, sexual function, bladder dysfunction, muscle spasms/spasticity, bowel dysfunction, circulatory problems, loss of sensation, urinary tract infection, respiratory problems, contractures, autonomic dysreflexia, postural hypertension, pressure injuries, and heterotopic ossification) utilizing the validated spinal cord injury-secondary conditions scale (30). Participants rated secondary conditions as none, infrequent, occasional, or chronically occurring problems. We dichotomized responses as none/infrequently/occasionally vs. chronically occurring problem. In addition, we created a count variable for the number of SCI/D secondary conditions (< 4 vs. 4+ SCI/D secondary conditions as a chronically occurring problem), for inclusion in the multivariable regression model.
Contributors to feelings of PSI were collected by an independent question asking Veterans to identify factors that contributed to their feelings of PSI at least some of the time. Participants were asked to select all that apply, the items were derived from qualitative themes identified through interviews with 33 Veterans with SCI/D focused on the most common factors that contributed to PSI in persons with SCI/D (LaVela 2022), options included: concerns about mobility, having a SCI/D, concerns about being a burden on others, environmental barriers, transportation issues/limitations, care needs due to injury (such as bowel care), limited social network, and concerns about people staring (31).
Analyses. Descriptive statistics were used to summarize demographics, injury characteristics, objective measures of social isolation, health factors, and contributors to feelings of PSI. Bivariate comparisons including chi-square and t tests, were used to assess unadjusted associations of high vs. low PSI for demographics, injury characteristics, objective measures of social isolation, and health factors.
Multivariable analysis. Multivariable logistic regression modeling was used to generate odds ratios (OR) and 95% confidence interval (CI) to determine variables independently associated with high PSI. Variables statistically significant in the bivariate analysis were considered for inclusion in the model. An alpha level of 0.05 was used to determine statistical significance.
Contributors to feelings of PSI. The independent question asking Veterans to identify contributors of PSI was only included in the descriptive analysis and not included in the bivariate or multivariable logistic regression, due to the dependent variable being measured in the question. We present percentages of endorsed responses.
Bias assessment. To assess non-response bias, we conducted bivariate comparisons of demographic and injury characteristics to determine if respondents (n = 421) differed from non-respondents (n = 1521). To assess self-selection bias, we conducted bivariate comparisons of demographic and injury characteristics to determine if our final model sample (n = 383) differed from those not in the model (n = 38) due to missing data. Statistical analyses were performed using SAS 9.4 (SAS Institute Inc., Cary, NC) and Stata version 17 software (StataCorp, College Station, TX).
Results
We invited 2000 Veterans to complete the survey, of those 58 Veterans were removed from the denominator for the following reasons: 54 undeliverable surveys, 2 deceased, 1 did not have a SCI/D, and 1 was in the intensive care unit. A total of 421 Veterans completed the survey (response rate of 22%, 421/1942). Analysis was conducted on data from 410 Veterans; 11 Veterans were excluded due to missing PSI data.
Demographics, injury characteristics, objective measures of social isolation, and health factors that could impact PSI are provided in Table 1. Over half (56%) had PSI mean scores higher than the general population. Most respondents were male (71%), white (76%) non-Hispanic (93%), married (56%), had paraplegia (50%), and had an incomplete (73%) traumatic SCI/D (60%). On average participants were 65 years old (range: 25–95), had injury duration of 18 years (range: 1–62), and experienced injury onset at 47 years old (range: 19–91). Bivariate findings show that participant characteristics associated with high vs low PSI included younger mean age (63 vs. 68 years P < 0.0001) and younger mean age at injury onset (45 vs. 49 years P = 0.0367). No differences were seen for sex, race, marital status, injury level, injury completeness, etiology, or duration of injury. Objective social isolation measures associated with high vs low PSI include those with no informal caregiver (39% vs. 29%, P = 0.036), fewer than 4 people in their social network (71% vs. 26%, P < 0.0001), infrequently or never having meaningful contacts (38% vs. 17%, P < 0.0001), and infrequently or never having reliable contacts (14% vs. 3%, P = 0.0003). There were no significant differences in high vs low PSI for living alone.
Table 1.
Demographic, injury characteristics, objective measures of social isolation, and health factors of Veterans with spinal cord injury and disorders by normal/low vs high perceived social isolation (PSI).
| n (N = 410) | Overall n (%) |
Low PSI: at or below population mean (< 50) n = 182 n (%) |
High PSI: above population mean (>50) n = 228 n (%) |
P value* | |
|---|---|---|---|---|---|
| Demographic characteristics | |||||
| Sex | |||||
| Male | 410 | 305 (74) | 143 (79) | 162 (71) | 0.0886 |
| Female | 105 (26) | 39 (21) | 66 (29) | ||
| Race | |||||
| White | 409 | 305 (75) | 133 (73) | 172 (76) | 0.7946 |
| Black | 86 (21) | 40 (22) | 46 (20) | ||
| Other | 18 (4) | 9 (5) | 9 (4) | ||
| Ethnicity | |||||
| Hispanic or Latino | 407 | 30 (7) | 14 (8) | 16 (7) | 0.8494 |
| Not Hispanic or Latino | 377 (93) | 166 (92) | 211 (93) | ||
| Marital Status | |||||
| Married | 410 | 243 (59) | 115 (63) | 128 (56) | 0.1578 |
| Never married, separated, widowed | 167 (41) | 67 (37) | 100 (44) | ||
| Age mean (sd) [range] | 410 | 65 (12) [25–95] | 68 (11) [34–95] | 63 (12) [25–93] | <0.0001 |
| Level of Injury | |||||
| Paraplegia | 397 | 203 (51) | 93 (53) | 110 (50) | 0.384 |
| Tetraplegia | 109 (27) | 42 (24) | 67 (30) | ||
| AIS D | 85 (21) | 40 (23) | 45 (20) | ||
| Completeness | |||||
| Complete | 280 | 71 (25) | 30 (23) | 41 (27) | 0.4925 |
| Incomplete | 209 (75) | 99 (77) | 110 (73) | ||
| Etiology | |||||
| Traumatic | 410 | 232 (69) | 114 (63) | 137 (60) | 0.599 |
| Non-traumatic | 105 (31) | 68 (37) | 91 (40) | ||
| Duration of injury mean (sd) [range] | 410 | 19 (15) [1–62] | 19 (16) [1–62] | 18 (15) [1–58] | 0.4892 |
| Age at injury mean (sd) [range] | 410 | 47 (17) [19–91] | 49 (18) [19–83] | 45 (16) [19–91] | 0.0367 |
| Objective measures of social isolation | |||||
| Informal Caregiver | |||||
| Informal Caregiver | 410 | 268 (65) | 129 (71) | 139 (61) | 0.036 |
| No informal caregiver | 142 (35) | 53 (29) | 89 (39) | ||
| Living arrangement | |||||
| Live alone | 406 | 110 (27) | 41 (23) | 69 (31) | 0.0735 |
| Does not live alone | 296 (73) | 140 (77) | 156 (69) | ||
| Social network size | |||||
| <4 people | 400 | 204 (51) | 46 (26) | 158 (71) | <0.0001 |
| 4+ people | 196 (49) | 133 (74) | 63 (29) | ||
| Frequency of meaningful contacts | |||||
| Hardly ever/never | 399 | 115 (29) | 30 (17) | 85 (38) | <0.0001 |
| Some of the time/Often | 284 (71) | 148 (83) | 136 (62) | ||
| Frequency of reliable contacts | |||||
| Hardly ever/never | 401 | 36 (9) | 6 (3) | 30 (14) | 0.0003 |
| Some of the time/Often | 365 (91) | 173 (97) | 192 (86) | ||
| Health Factors | |||||
| Complications last 6 months | |||||
| Depression | 410 | 206 (50) | 37 (20) | 169 (74) | <0.0001 |
| Anxiety | 410 | 197 (48) | 38 (21) | 159 (70) | <0.0001 |
| Dysfunctional sleep | 410 | 245 (60) | 82 (45) | 163 (71) | <0.0001 |
| Post-traumatic stress | 410 | 146 (36) | 27 (15) | 119 (52) | <0.0001 |
| Number of secondary conditions identified as a chronic problem** | |||||
| <4 Chronic | 401 | 226 (56) | 123 (69) | 103 (46) | <0.0001 |
| ≥4 Chronic | 175 (44) | 54 (31) | 121 (54) | ||
| Individual secondary conditions*** | |||||
| Chronic pain | |||||
| None/Infrequent/Occasional | 388 | 155 (40) | 93 (55) | 62 (28) | <0.0001 |
| Chronic | 233 (60) | 76 (45) | 157 (72) | ||
| Joint and muscle pain | |||||
| None/Infrequent/Occasional | 391 | 185 (47) | 107 (63) | 78 (35) | <0.0001 |
| Chronic | 206 (53) | 63 (37) | 143 (65) | ||
| Sexual dysfunction | |||||
| None/Infrequent/Occasional | 380 | 185 (49) | 89 (54) | 96 (45) | 0.0904 |
| Chronic | 195 (51) | 77 (46) | 118 (55) | ||
| Bladder dysfunction | |||||
| None/Infrequent/Occasional | 385 | 217 (56) | 108 (65) | 109 (50) | 0.0040 |
| Chronic | 168 (44) | 59 (35) | 109 (50) | ||
| Muscle spasms/spasticity | |||||
| None/Infrequent/Occasional | 376 | 234 (62) | 116 (71) | 118 (55) | 0.0018 |
| Chronic | 142 (38) | 47 (29) | 95 (45) | ||
| Bowel dysfunction | |||||
| None/Infrequent/Occasional | 381 | 246 (65) | 122 (74) | 124 (58) | 0.0014 |
| Chronic | 135 (35) | 44 (27) | 91 (42) | ||
| Circulatory problems | |||||
| None/Infrequent/Occasional | 361 | 283 (78) | 130 (84) | 153 (74) | 0.0283 |
| Chronic | 78 (22) | 25 (16) | 53 (26) | ||
| Loss of sensation | |||||
| None/Infrequent/Occasional | 359 | 306 (85) | 135 (85) | 171 (86) | 0.8747 |
| Chronic | 53 (15) | 24 (15) | 29 (15) | ||
| Urinary tract infection(s) | |||||
| None/Infrequent/Occasional | 378 | 327 (87) | 149 (91) | 178 (83) | 0.0304 |
| Chronic | 51 (13) | 15 (9) | 36 (17) | ||
| Respiratory problems | |||||
| None/Infrequent/Occasional | 368 | 332 (90) | 152 (94) | 180 (87) | 0.0387 |
| Chronic | 36 (10) | 10 (6) | 26 (13) | ||
| Contractures | |||||
| None/Infrequent/Occasional | 322 | 290 (90) | 135 (94) | 155 (87) | 0.0466 |
| Chronic | 32 (10) | 9 (6) | 23 (13) | ||
| Autonomic dysreflexia | |||||
| None/Infrequent/Occasional | 323 | 294 (91) | 140 (95) | 154 (88) | 0.0154 |
| Chronic | 29 (9) | 7 (5) | 22 (13) | ||
| Postural hypertension | |||||
| None/Infrequent/Occasional | 315 | 294 (93) | 134 (96) | 160 (91) | 0.1297 |
| Chronic | 21 (7) | 6 (4) | 15 (9) | ||
| Pressure injuries | |||||
| None/Infrequent/Occasional | 354 | 334 (94) | 146 (94) | 188 (95) | 0.5822 |
| Chronic | 20 (6) | 10 (6) | 10 (5) | ||
| Heterotopic ossification | |||||
| None/Infrequent/Occasional | 312 | 296 (95) | 133 (94) | 163 (95) | 0.6916 |
| Chronic | 16 (5) | 8 (6) | 8 (5) |
*chi-square tests were performed on all variables except for the continuous variables age, duration of injury, and age at injury for which t tests were conducted.
**≥4 SCI secondary conditions that were a chronic problem vs. <4 secondary conditions that were a chronic problem (vs. none, infrequent, or mild).
*** Presented in order of overall column highest to lowest percentage of secondary conditions reported as chronic problems.
The most reported complication in the last 6 months was dysfunctional sleep (60%). Complications in the last 6 months associated with high vs low PSI include having depression (74% vs. 20%, P < 0.0001), anxiety (70% vs. 21%, P < 0.0001), dysfunctional sleep (71% vs 45%, P < 0.0001), and post-traumatic stress (52% vs 15%, P < 0.0001) in the last 6 months. Most reported secondary conditions were chronic pain (72%), joint/muscle pain (65%), and sexual dysfunction (55%), and bladder dysfunction (50%). Bivariate comparisons showed statistically significant high PSI vs low PSI for all SCI/D secondary conditions except sexual dysfunction, loss of sensation, postural hypertension, pressure injuries, and heterotopic ossification. In addition, a greater proportion of individuals had 4 or more SCI/D secondary conditions that were chronically occurring problems (high PSI (54%) vs. low PSI (31%), P < 0.0001).
The final multivariable logistic regression model for variables associated with PSI is presented in Table 2. Among the objective measures, having a smaller social network size was associated with increased odds of high PSI (OR 3.59, P < .0001). Among health factors, having depression (OR 3.98, P < .0001), anxiety (OR 2.29, P = 0.009), and post-traumatic stress (OR 2.56, P = 0.003) in the previous 6 months and having 4 or more SCI/D secondary conditions be chronically occurring problems (OR 1.78, P = 0.045) was associated with increased odds of high PSI.
Table 2.
Multivariable logistic regression: Variables independently associated with high perceived social isolation [reference: normative/low perceived social isolation] (n = 383).
| Variables | Odds Ratio | 95% Confidence interval | P value |
|---|---|---|---|
| Age (years) | 0.991 | 0.964–1.02 | 0.546 |
| Age at injury (years) | 0.999 | 0.971–1.01 | 0.225 |
| Objective measures of social isolation | |||
| Informal caregiver | |||
| Informal Caregiver [ref: no informal caregiver] | 0.570 | 0.317–1.03 | 0.061 |
| Social network size | |||
| <4 people [ref: 4+] | 3.59 | 2.07–6.21 | <0.0001 |
| Frequency of meaningful contacts | |||
| Hardly ever/never [ref: some of the time/often] | 1.58 | 0.828–3.00 | 0.166 |
| Frequency of reliable contacts | |||
| Hardly ever/never [ref: some of the time/often] | 3.08 | 0.953–9.92 | 0.060 |
| Health factors | |||
| Depression | 3.98 | 2.13–7.40 | <0.0001 |
| Anxiety | 2.29 | 1.23–4.30 | 0.009 |
| Dysfunctional sleep | 0.969 | 0.532–1.76 | 0.918 |
| Post traumatic stress | 2.56 | 1.38–4.77 | 0.003 |
| Secondary conditions identified as a chronic problem | |||
| ≥4 secondary conditions [ref: <4] | 1.78 | 1.01–3.11 | 0.045 |
Contributors to feelings of PSI. Self-identified contributors to PSI are shown in Fig. 1. Almost two thirds of Veterans identified mobility (63%), and over half indicated having SCI/D (61%) and concerns about being a burden on others (57%) as contributing to feelings of PSI.
Figure 1.
Percent of veterans with spinal cord injury/disorders (SCI/D) identified contributors of perceived social isolation*. *Questions were select all that apply; n = 410.
Non-response bias. There were no statistically significant differences for respondents (n = 421) vs non-respondents (n = 1521) for race, ethnicity, duration of injury, and level of injury. There were statistically significant differences between respondents andnon-respondents for age [mean age 65 years for respondents vs 59 years for non-respondents, P < 0.0001], age at injury [mean age 47 for respondents vs 43 for non-respondents, P < 0.0001] and sex [26% of respondents where female vs 21% of non-respondents, P = 0.045].
Selection bias. There were no differences in most variables, including PSI, demographics, and injury characteristics. The only exceptions of persons included in the regression model (n = 383) vs. those not in the model (n = 38) were marital status [61% in the model indicated married vs 37% not in the model, P = 0.004] and level of injury [52% in the model indicated paraplegia vs 38% not in the model; 28% in the model indicated tetraplegia vs 24% not in the model; and 21% in the model indicated AIS D vs 38% not in the model, P = 0.05].
Discussion
Our analysis aimed to describe PSI among Veterans living with SCI/D, specifically, and identify which demographic, injury characteristics, objective social isolation measures, and health factors were significantly associated with high PSI. Overall, PSI was prevalent in over half of the sample of Veterans living with SCI/D. Our findings further indicated that objective social isolation measures including social network size, health factors including depression, anxiety, and post-traumatic stress, and multiple chronically occurring SCI/D secondary conditions were statistically associated with high PSI.
We found high PSI in Veterans with SCI/D who had less than 4 people in their social network. Although network size and quality of interactions are reduced following a SCI/D (2), there is an association between small social network and PSI in this population (1). However, other cross-sectional surveys of individuals living with SCI/D did not find an association with PSI and smaller network size but found associations with fewer number of contacts and lower levels of intimacy with contact (2, 32). In our survey we measured a similar concept, the frequency of reliable contacts (e.g. how often can you rely on others if you have a problem?) and we found that hardly ever/never having someone to rely on was statistically related to high PSI in the bivariate analysis and the multivariable analysis showed a trend for 3 times higher odds of high PSI (P = 0.06). In addition, we measured the composition of social network by analyzing the presence or absence of an informal caregiver. The presence of an informal caregiver was found to have lower odds, or a protective effect (P = 0.06) of having high PSI in the multivariable analysis and the bivariate analysis showed those with an informal caregiver reported greater frequency of low PSI (P = 0.036). There are mixed findings on how having an informal caregiver can affect PSI. Some studies looking at informal caregivers of individuals with SCI/D share similar findings as our study that informal caregivers help decrease PSI, by giving individuals higher perceived social support (15, 33). However, other studies have found that once the role of a family member, spouse/partner, or friend change to that of an informal caregiver the relationship/intimacy dynamic changes and negatively impacts PSI (34). Our findings suggest that measuring social network size represents an important indicator of PSI. Additionally, social support which is a construct that may be related to one’s social network is associated with better health outcomes in individuals with SCI/D (14). These findings suggest that social network size and level of support may be beneficial to target in an intervention to lower PSI. However, more research in Veterans living with SCI/D is warranted to understand the types of social network needed, such as family, friends, Veterans living with SCI/D acting as peer support or role models, caregivers, and other type of social connections that are the most beneficial to lowering PSI. Future research is needed to examine what kinds of social support (i.e. emotional support, instrumental support, or information support) are related to PSI and health outcomes in Veterans living with SCI/D.
Our findings showed that higher PSI was associated with three distinct mental health problems, depression, anxiety, and post-traumatic stress. Our results align with literature indicating that PSI is associated with developing, having, or worsening of depression and anxiety (22, 32, 35). In a systematic review of the general population, adults who reported feeling lonely more often had more than double the odds of developing depression compared to those who rarely/never feel lonely (36). Our results indicate those with depression having almost 4 times greater odds of higher PSI vs those without depression is consistent with the general population findings (36). Additionally, a study focusing on individuals with SCI/D found that increasing PSI, also increased depression (r = .29, P < .001; reflecting a medium effect size) (37). Our study also found high PSI was associated with post-traumatic stress, which is similar to other studies such as a systemic review of loneliness among Veterans found loneliness to be related to symptoms of post-traumatic stress disorder (38) and a longitudinal cohort study found greater hazard for post-traumatic stress disorder among individuals with SCI/D compared to those without a SCI/D. Further research is needed to better understand the relationship between PSI and mental health conditions such as depression, anxiety, and post-traumatic stress, the underlying mechanisms, and what potential interventions addressing these mental health conditions would be on PSI.
In addition, having four or more chronically occurring SCI/D secondary conditions was associated with nearly 2 times greater odds of high PSI. A scoping review of individuals living with SCI/D found associations between loneliness and loss of sexual function, bladder disorders, and symptoms of SCI/D (1). This is similar to our findings, with some of the most common significant SCI/D secondary conditions identified in our study being pain, bladder dysfunction, and muscle spasms/spasticity, although sexual dysfunction was shown to be higher in those with high PSI it was not significant. It would be beneficial to identify the ways in which secondary conditions in Veterans living with SCI/D contribute to PSI in order to develop interventions to improve SCI/D secondary condition management in ways that would alleviate PSI.
Respondents identified concerns they felt most commonly impacted their PSI were related to their SCI/D diagnosis. In other words, they felt that simply having a SCI/D influenced their feelings of PSI. Participants identified mobility as a common concern impacting their PSI. Robinson-Whelen et al. (39) similarly found mobility to be associated with loneliness in individuals living with SCI/D. A qualitative study looking at factors associated with PSI (40) also found that mobility was a bigger issue due to environmental barriers, such as when a building lacked accessibility and during winter months due to weather conditions making it more difficult for social interactions. In the same study, individuals also described feeling like they were a burden due to their SCI/D care and needs, which is similar to our study’s findings of having concerns about being a burden on others as a contributor of PSI. Both mobility impairments and the feeling of being a burden could cause individuals to feel disconnected from individuals socially, which can affect factors of PSI such as engagement with social relationships or maladaptive social cognition, i.e. negative thoughts about self-worth (41). Interventions, such as cognitive behavioral therapy, addressing maladaptive ways of thinking been reported to be effective in reducing PSI (41, 42). It is important to statistically identify what is associated with high PSI, but it is also essential to identify what the individual feels in contributory (42), such as impaired mobility and feeling like a burden.
Limitations: The analysis was limited by potential selection bias given the 22% survey response rate. A cross-sectional survey design with self-reported responses could have recall bias affecting responses. While there were no differences in the majority of variables, our assessments of potential sources of bias indicated that there was a respondent bias towards older persons and women and a model selection bias towards those who were married, people with paraplegia, and those with AIS D injury severity. These biases may impact the generalizability of the findings. Finally, Veterans with SCI/D may differ from other SCI/D cohorts which may limit generalizability.
Conclusion
Objective social isolation measures such as social network size may be used to identify Veterans with SCI/D at risk for PSI, and more research for future interventions needs to be conducted on the type and quantity of social network interactions are most beneficial in reducing PSI in this population. A better understanding of the co-occurrence of mental health conditions such as such as depression, anxiety, and post-traumatic stress is needed to understand if one precedes the other or if addressing one improves the other. In addition, research is warranted on the impact of having chronically occurring SCI/D secondary conditions and how care needs can impact social health, e.g. bowel care programs may stifle social plans. In addition, contributors identified by Veterans with SCI/D such as “feeling like a burden to others” may benefit from different interventions such as targeting maladaptive social cognition. These findings reinforce the complexity of PSI and the need for more research to develop patient-centered interventions aimed at reducing PSI. These findings help in identifying what variables are associated with high PSI in Veterans with SCI/D, which will aid in the creation of evidence-based interventions for PSI as requested in the Surgeon General’s advisory on the epidemic of loneliness and social isolation in the U.S.
Acknowledgements
The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans’ Affairs or the United States government.
Funding Statement
This study was supported by the Paralyzed Veterans of America Education [grant number 876], “Social Isolation and Loneliness in Persons with SCI/D: Developing an Educational Curriculum for Health Care Professionals” [PI: LaVela, SL].
Disclaimer statements
Contributors None.
Conflicts of interest Authors have no conflict of interests to declare.
References
- 1.Cimino SR, Cadel L, Guilcher SJT, Wasilewski M, Hitzig SL.. Social disconnectedness and perceived social isolation in persons with spinal cord injury/dysfunction living in the community: a scoping review. J Spinal Cord Med. 2023;46(3):367–389. doi: 10.1080/10790268.2022.2129170. Epub ahead of print. PMID: 36269865. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Guilcher SJ, Craven B, Bassett-Gunter RL, Cimino SR, Hitzig SL.. An examination of objective social disconnectedness and perceived social isolation among persons with spinal cord injury/dysfunction: a descriptive cross-sectional study. Disabil Rehabil. 2021;43(1):69–75. [DOI] [PubMed] [Google Scholar]
- 3.Holt-Lunstad J, Smith TB, Baker M, Harris T, Stephenson D.. Loneliness and social isolation as risk factors for mortality: a meta-analytic review. Perspect Psychol Sci. 2015;10(2):227–237. [DOI] [PubMed] [Google Scholar]
- 4.Newman SD, Li C, Krause JS.. Social isolation after spinal cord injury: indicators from the longitudinal aging study. Rehabil Psychol. 2016;61(4):408–416. [DOI] [PubMed] [Google Scholar]
- 5.Tate DG, Forchheimer M, Reber L, Meade M, Tan N, Clarke P.. Factors enabling comorbidities and secondary conditions in older adults with spinal cord injury. J Spinal Cord Med. 2023;46(6):929–940. doi: 10.1080/10790268.2022.2108662. Epub ahead of print. PMID: 35993788. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Umberson D, Karas Montez J.. Social relationships and health: a flashpoint for health policy. J Health Soc Behav. 2010;51(1_suppl):S54–S66. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Havens B, Hall M, Sylvestre G, Jivan T.. Social isolation and loneliness: differences between older rural and urban Manitobans. Can J Aging. 2004;23(2):129–140. [DOI] [PubMed] [Google Scholar]
- 8.Nicholson NR Jr. Social isolation in older adults: an evolutionary concept analysis. J Adv Nurs. 2009;65:1342–1352. [DOI] [PubMed] [Google Scholar]
- 9.Fried L, Prohaska T, Burholt V, Burns A, Golden J, Hawkley L, Lawlor B, Leavey G, Lubben J, O’Sullivan R, et al. A unified approach to loneliness. Lancet. 2020;395(10218):114. 10.1016/S0140-6736(19)32533-4 [DOI] [PubMed] [Google Scholar]
- 10.Coyle CE, Dugane E.. Social isolation, loneliness and health among older adults. J Aging Health. 2012;24(8):1346–1363. [DOI] [PubMed] [Google Scholar]
- 11.Nemecek D. Cigna US Loneliness Index: survey of 20,000 Americans examining behaviors driving loneliness in the United States. Bloomfield (CT: ): Cigna; 2018. [Google Scholar]
- 12.Emerson E, Fortune N, Llewellyn G, Stancliffe R.. Loneliness, social support, social isolation and wellbeing among working age adults with and without disability: cross-sectional study. Disabil Health J. 2021;14(1):100965. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Tough H, Brinkhof MW, Siegrist J, Fekete C.. The impact of loneliness and relationship quality on life satisfaction: a longitudinal dyadic analysis in persons with physical disabilities and their partners. J Psychosom Res. 2018;110:61–67. [DOI] [PubMed] [Google Scholar]
- 14.Müller R, Peter C, Cieza A, Geyh S.. The role of social support and social skills in people with spinal cord injury—a systematic review of the literature. Spinal Cord. 2012;50(2):94–106. [DOI] [PubMed] [Google Scholar]
- 15.Barclay L, McDonald R, Lentin P, Bourke-Taylor H.. Facilitators and barriers to social and community participation following spinal cord injury. Aust Occup Ther J. 2016;63(1):19–28. [DOI] [PubMed] [Google Scholar]
- 16.Straus E, Norman SB, Tripp JC, Tsai J, Sippel LM, Jeste DV, Pietrzak RH.. Behavioral epidemic of loneliness in older US military veterans: results from the 2019–2020 national health and resilience in veterans study. Am J Geriatr Psychiatry. 2022;30(3):297–310. [DOI] [PubMed] [Google Scholar]
- 17.National Spinal Cord Injury Statistical Center . Traumatic Spinal Cord Injury Facts and Figures at a Glance. Birmingham (AL). Available from: https://www.nscisc.uab.edu/public/Facts%20and%20Figures%202023%20-%20Final.pdf 2023.
- 18.Sippel JL, Daly JE, Poggensee L, Ristau KD, Eberhart AC, Tam K, Smith BM.. Modernization of a large spinal cord injuries and disorders registry: the veterans administration experience. Arch Rehabil Res Clin Transl. 2022;4(4):100237. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Shen C, Rolls ET, Cheng W, Kang J, Dong G, Xie C, Zhao XM, Sahakian BJ, et al. Associations of social isolation and loneliness with later dementia. Neurology. 2022;99(2):e164–e175. [DOI] [PubMed] [Google Scholar]
- 20.Jiang HX, Liu Y, Jiang JJ, Wei JH, Niu CC, Yu J.. Community support, social isolation and older adults’ life satisfaction: evidence from a national survey in China. Aging Ment Health. 2023: 1–9. Epub ahead of print. PMID: 37395111. [DOI] [PubMed] [Google Scholar]
- 21.Furman D, Campisi J, Verdin E, Carrera-Bastos P, Targ S, Franceschi C, Slavich GM.. Chronic inflammation in the etiology of disease across the life span. Nat Med. 2019;25(12):1822–1832. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Smith BJ, Lim MH.. How the COVID-19 pandemic is focusing attention on loneliness and social isolation. Public Health Res Pract. 2020;30:3022008. doi: 10.17061/phrp3022008. [DOI] [PubMed] [Google Scholar]
- 23.Cene CW, Beckie TM, Sims M, Suglia SF, Aggarwal B, Moise N, McCullough LD.. American Heart Association Social Determinants of Health Committee of the Council on Epidemiology and Prevention and Council on Quality of Care and Outcomes Research; et al. Effects of Objective and Perceived Social Isolation on Cardiovascular and Brain Health: A Scientific Statement From the American Heart Association. J Am Heart Assoc. 2022;11:e026493. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Kanbay M, Tanriover C, Copur S, Peltek IB, Mutlu A, Mallamaci F, Zoccali C.. Social isolation and loneliness: undervalued risk factors for disease states and mortality. Eur J Clin Investig. 2023: e14032. [DOI] [PubMed] [Google Scholar]
- 25.United States. Public Health Service. Office of the surgeon general. Our epidemic of loneliness and isolation: A report of the surgeon general. U.S. Department of Health and Human Services, Public Health Service, Office of the Surgeon General 2023.
- 26.LaVela SL, Etingen B, Miskevics S, Heinemann AW.. What determines low satisfaction with life in individuals with spinal cord injury? J Spinal Cord Med. 2019;42(2):236–244. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.LaVela SL, Heinemann AW, Etingen B, Miskovic A, Locatelli SM, Chen D.. Relational empathy and holistic care in persons with spinal cord injuries. J Spinal Cord Med. 2017;40(1):30–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Cella D, Riley W, Stone A, Rothrock N, Reeve B, Yount S, Hays R.. The patient-reported outcomes measurement information system (PROMIS) developed and tested its first wave of adult self-reported health outcome item banks: 2005–2008. J Clin Epidemiol. 2010;63(11):1179. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Questionnaires | Health and Retirement Study [Internet]. hrs.isr.umich.edu. Available from: https://hrs.isr.umich.edu/documentation/questionnaires.
- 30.Jørgensen V, Von Rosen P, Butler Forslund E.. Considerations on the psychometric properties and validity of the spinal cord injury secondary conditions scale. Spinal Cord. 2021;59(8):894–901. 10.1038/s41393-021-00655-z. [DOI] [PubMed] [Google Scholar]
- 31.LaVela SL, et al. Academy of spinal cord injury professionals 2022 ASCIP hybrid educational conference & expo September 7–10 Kansas City, MO, platform abstracts 1–23. J Spinal Cord Med. 2022;45(5):773–788. [Google Scholar]
- 32.Hitzig SL, Cimino SR, Alavinia M, Bassett-Gunter RL, Craven BC, Guilcher SJ.. Examination of the relationships among social networks and loneliness on health and life satisfaction in people with spinal cord injury/dysfunction. Arch Phys Med Rehabil. 2021;102(11):2109–2116.e1. [DOI] [PubMed] [Google Scholar]
- 33.Jensen MP, Smith AE, Bombardier CH, Yorkston KM, Miró J, Molton IR.. Social support, depression, and physical disability: age and diagnostic group effects. Disabil Health J. 2014;7(2):164–172. [DOI] [PubMed] [Google Scholar]
- 34.Quinn C, Clare L, Woods B.. The impact of the quality of relationship on the experiences and wellbeing of caregivers of people with dementia: a systematic review. Aging Ment Health. 2009;13(2):143–154. [DOI] [PubMed] [Google Scholar]
- 35.Domènech-Abella J, Mundó J, Haro JM, Rubio-Valera M.. Anxiety, depression, loneliness and social network in the elderly: longitudinal associations from The Irish Longitudinal Study on Ageing (TILDA). J Affect Disord. 2019;246:82–88. [DOI] [PubMed] [Google Scholar]
- 36.Mann F, Wang J, Pearce E, Ma R, Schlief M, Lloyd-Evans B. Loneliness and the onset of new mental health problems in the general population: a systematic review. medRxiv. [Preprint]. 2021.
- 37.Robinson-Whelen S, Hughes RB, Taylor HB, Holmes SA, Rodriquez J, Manohar S.. Impacts of the COVID-19 pandemic on people with spinal cord injury. Rehabil Psychol. 2023;68(1):1. [DOI] [PubMed] [Google Scholar]
- 38.Wilson G, Hill M, Kiernan MD.. Loneliness and social isolation of military veterans: systematic narrative review. Occup Med (Chic Ill). 2018;68(9):600–609. [DOI] [PubMed] [Google Scholar]
- 39.Robinson-Whelen S, Taylor HB, Feltz M, Whelen M.. Loneliness among people with spinal cord injury: exploring the psychometric properties of the 3-item loneliness scale. Arch Phys Med Rehabil. 2016;97(10):1728–1734. [DOI] [PubMed] [Google Scholar]
- 40.Cimino SR, Hitzig SL, Craven BC, Bassett-Gunter RL, Li J, Guilcher SJ.. An exploration of perceived social isolation among persons with spinal cord injury in Ontario, Canada: a qualitative study. Disabil Rehabil. 2022;44(14):3400–3409. [DOI] [PubMed] [Google Scholar]
- 41.Mann F, Bone JK, Lloyd-Evans B, Frerichs J, Pinfold V, Ma R, Johnson S.. A life less lonely: the state of the art in interventions to reduce loneliness in people with mental health problems. Soc Psychiatry Psychiatr Epidemiol. 2017;52:627–638. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Masi CM, Chen HY, Hawkley LC, Cacioppo JT.. A meta-analysis of interventions to reduce loneliness. Pers Soc Psychol Rev. 2011;15(3):219–266. [DOI] [PMC free article] [PubMed] [Google Scholar]