Introduction

People with spinal cord injury (SCI) face challenges in their everyday lives including limitations in functioning, secondary conditions, medical complications and decreased quality of life.1, 2 Social support represents an important resource to meet these challenges. Social support is known to positively influence health,3, 4, 5, 6 life satisfaction7, 8 and even mortality.9, 10 Social support is crucial, but can be diminished after SCI.11

Social support is defined as an exchange of resources between individuals intended to enhance the well-being of the recipient.12 It conveys the information of being loved, cared for, esteemed, valued and bestows a sense of belonging.13 Social support can be instrumental (such as tangible assistance), emotional (such as exchange with a close friend) or informational (such as advice from a peer). Social support can be described from a quantitative (for example, network size) or qualitative (for example, satisfaction with support) perspective.14, 15, 16, 17

It has been shown that mobilization of social support can be influenced by social skills.18, 19, 20, 21 Social skills, according to evolutionary theory, are prerequisites for survival and adaptation.22 Depression,21, 23, 24 social phobia,25 substance or alcohol abuse,26, 27 adherence in rehabilitation,28 life satisfaction and quality of life20 correlate with social skill deficits. Social skills are important for people with disability. They help to overcome discomfort and stigmatization in social situations, to ask for help, to put others at ease in social relationships, to elicit feedback and to develop and foster social relationships.29, 30

Social skills are defined as the ability to interact with other people in a manner that is both appropriate and effective.31 They comprise aspects of verbal and non-verbal communication. They include, for example, styles of social problem solving (such as rational, impulsive or avoidant), confidence, assertiveness, goal direction or self-monitoring.32

There are published reviews on social support in coronary heart disease,33 stroke,34 cancer35 or diabetes.36 There are also existing reviews on social skills in depression,23, 24 schizophrenia37, 38 and autism.39, 40 In the field of SCI, Kreuter11 has reviewed the literature regarding partner relationships. However, a comprehensive overview on social support or social skills in SCI does not yet exist.

Therefore, the aim of this study is to examine the current knowledge of how social support and social skills are associated with aspects of health, functioning and quality of life of persons living with SCI. The specific aims are to answer the questions (1) which aspects of social support and social skills are addressed in SCI research, (2) which methods are used to assess social support and social skills and (3) to summarize the evidence about social skills and social support in SCI.

Methods

A systematic literature review was conducted to identify scientific publications which refer to social support and social skills in people with SCI. The procedures followed five steps: electronic literature search, paper selection, data extraction, quality assessment of the studies and narrative synthesis.

Searches were conducted in Pubmed, Embase, PsycINFO, ERIC (Educational Resources Information Centre), CINAHL (Cumulative Index to Nursing and Allied Health Literature) and the SSCI (Social Sciences Citation Index). The search terms ‘social support’, ‘social skills’ and synonyms combined with ‘spinal cord injury’ were used.

Publications were selected that generate data which target, assess and intervene in social support and/or social skills. Studies referring to different dimensions of social support, that is, type (such as emotional, instrumental, informational), source (such as family, friends, peers) and qualifier (such as satisfaction, appreciation with social support) were selected. Support that is paid, such as professional support, was excluded. The theoretical framework of Liberman32 was used to capture the multidimensional types of social skills (that is, topographical, functional, information processing). The topographical dimension emphasizes on verbal and non-verbal behavior (such as communication skills, eye contact). The functional view defines social skills in terms of the outcome of social interactions (such as assertiveness, self-monitoring). Information-processing skills refer to the individual's ability to attend to, receive, process cues, generate and decide on a response and implement it (for example, social problem solving, decision making). Randomized controlled or clinical trials, cross-sectional or longitudinal studies, published in English between 1990 and 2010 with a sample of people with SCI, who are at least 13 years of age were selected. In addition, studies with a sample size smaller than 30, qualitative and psychometric studies, reviews, meta-analyses and studies in which SCI was not the main target population were excluded.

Data extraction included documentation of the main objective, study design, country, size and description of the sample. In addition, the variables assessed, measurement instruments and the results of the study were extracted.

For quality assessment of the studies, evidence grading according to STROBE (Strengthening the Reporting of Observational Studies in Epidemiology)41 and PEDro (Physiotherapy Evidence Database) (http://www.pedro.org.au/english/downloads/pedro-scale/ (cited 30 July 2010)) were implemented. STROBE represents a quality assessment tool for observational studies, which consists 22 items to evaluate the background, study design, data collection and data analysis of the study. PEDro includes 11 criteria, such as randomization, concealed allocation, blinding, etc., relevant for randomized controlled trials. The criteria fulfilled by STROBE and PEDro were counted.

Finally, results about social support and social skills were grouped according to topic domains, which represent the variables in relation to which social support and social skills have been studied. The narrative synthesis42 considered the number of studies pertaining to a topic domain, the statistical significance and consistency of the results, the analyses methods and the methodological quality of the study, including design, sample size, application of standardized measures or potential sources of bias.

For quality assurance, paper selection, data extraction (for one-third of the publications) and quality assessment of the studies were conducted in parallel by two independent reviewers. To resolve disagreements between the two reviewers, the original paper was consulted and rating mistakes, if any, were corrected. In case of controversial issues, a discussion was led by a third person, in which the two reviewers stated their pros and cons for the decision regarding paper selection, data extraction or quality assessment. On the basis of these statements, the third person made an informed decision. All review steps were conducted using an MS-Access database(Access 2007, Microsoft Corporation, Redmond, WA, USA.).

Results

The electronic searches in the 6 databases resulted in 795 hits. In all, 58 papers on social support, 11 on social skills and 1 study including both constructs were eligible for analyses (Figure 1). Study characteristics, demographical and lesion-related data of the study populations are summarized in Table 1. The majority of the papers were cross-sectional studies (n=44). Most studies were conducted in the United States (n=32). Sample sizes ranged between 33 and 1312. Two-thirds of the participants were male. Paraplegia and tetraplegia, as well as complete and incomplete lesions were approximately equally distributed.

Figure 1
figure 1

Flowchart of the systematic literature review.

Full size image
Table 1 Characteristics of the 69 papers included about social support (n=58) and social skills (n=11)
Full size table

Reviewer agreement on paper selection was 81%. On data extraction of variables and measurement instruments, agreement was 82%, agreement on results was 81% and agreement for STROBE quality assessment was 94%.

Table 2 shows the various aspects of social support addressed in SCI research. Studies focused on emotional (n=9), instrumental (n=9) and informational (n=9) aspects of social support provided by family (n=8), friends (n=8), intimate partners (n=8), peers (n=1) and community (n=1). Quality, that is, satisfaction with social support (n=9) and quantity of social support, for example, numbers of friends (n=9), were captured. Table 3 shows the four different social skills examined in persons with SCI: social problem-solving ability (n=7), assertiveness (n=3), self-monitoring (n=1) and communication skills (n=1).

Table 2 Social support variables and measurement instruments extracted from 58 papers
Full size table
Table 3 Social skill variables and measurement instruments extracted from 11 papers
Full size table

In all, 14 standardized self-report instruments assessing social support were used in 58 studies (Table 4). The most commonly used instrument in SCI was the ISEL (Interpersonal Support Evaluation List),43 measuring the availability of different types of social support. Five standardized self-report instruments were used to assess social skills (Table 5). As social problem solving is the most frequently examined social skill in SCI, the Social Problem-Solving Inventory—Revised,57 assessing problem orientation and problem-solving skills, was most commonly used.

Table 4 Self-report standardized instruments measuring social support extracted from 58 papers
Full size table
Table 5 Self-report standardized instruments measuring social skills extracted from 11 papers
Full size table

With regard to study quality, percentage scores on the STROBE ranged from 50.0 to 86.4%. Figure 2 shows the histogram of the results, demonstrating a normal distribution located in the upper half of possible percentage scores (mean=68%; range=50–86%; s.d.=8.76). The quality assessment according to PEDro was used in one study (N=40), which scored 7 out of 1161. Considering the methodological characteristics of the studies, the strengths of evidence is frequently diluted, because most of the results referred to bivariate correlations, which cannot specify direction or causal mechanisms of relationships. In addition, because of the lack of representativeness of the samples, the results of the identified studies cannot be generalized.

Figure 2
figure 2

Distribution of the STROBE results.

Full size image

Social skills and social support

Only one cross-sectional study (N=156) addressed the relationship of social support to social skills.62 Correlations between assertiveness and different types of social support were non-significant (r=0.13 to −0.38). However, including the interaction between assertiveness and social support in a multivariate analysis (together with sociodemographic and injury-related variables) revealed an association with depression and psychosocial disability. It indicates that assertive people were found to be more depressed and psychosocially disabled under conditions of high informational support. The model of the relationship between assertiveness and social support accounted for 38% of the variance in depression scores.

Social support

Findings about social support were grouped into eight topical domains (Table 6, Figure 3).

Table 6 Summary results on the associations of social support with aspects of health, functioning and quality of life
Full size table
Figure 3
figure 3

Overview of the systematic literature review results.

Full size image

Mental health

The most consistent relationship identified in this review is that between social support and mental health. This is due to the large number of studies that report significant associations between them. In 16 studies (N=33–256), social support was associated with lower depression,62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 107, 108 helplessness,75 pessimism,68 negative thoughts about the world and about oneself,76, 77 alexithymia77 and suicidal ideation.68, 78 In one study, social support accounted for 26.5% of the variance in hopelessness.68 In four studies (N=37–165) social support correlated with anxiety and moderated the relation between stress and anxiety.61, 63, 65, 71 Social support was related to less psychosocial disability (N=156–290)62, 79, 80 and lower severity of post-traumatic stress disorder (N=50–168) in one longitudinal study.76, 81, 82, 83 Alcohol and drug use ideation was associated with lower quality of social support,71 and pre-injury drinker reported lower levels of social support and perceived higher family than friend support.84

Life satisfaction, subjective well-being and quality of life

Evidence of the relation between social support and life satisfaction, quality and well-being is consistent, showing similar results in different studies. In all, 12 studies (N=62–256) showed that social support was associated with life satisfaction, subjective well-being and quality of life.64, 65, 69, 72, 74, 80, 85, 86, 87, 88, 89, 90 However, lower satisfaction with social life was associated with higher instrumental and informational support, higher emotion-oriented support from friends and lower from family.86 The availability of peer support positively affected satisfaction with life.74

Mortality, morbidity, secondary conditions and health-care utilization

Evidence regarding social support and association with mortality is mainly supported by a longitudinal study (N=1312) on survival analysis. The mortality risk decreased by 14% with every s.d. unit increase in reciprocal social support.91 With respect to morbidity, 11 studies showed correlations between social support and better health (N=125–475),72, 74, 85, 86, 89, 90, 92 lower frequency of health problems,80 disability-related problems80 and secondary conditions,93 such as urinary tract infections47, 72 and pressure ulcers.94, 47 Three studies showed inconsistent results in the relation between social support and numbers of days in hospital, hospital admissions and doctor visits, depending on the kind and source of social support.47, 70, 87 Emotional support was positively linked to health-care use,93 indicating that the more social support, the more health-care use.

Pain

In four cross-sectional studies (N=96–182), correlations between social support and lower degree of pain and catastrophizing were found.47, 92, 95, 96 However, the direction of the relationship cannot be determined. Informational and instrumental support were positively related to pain, and negatively related to emotional support.47

Beliefs, coping and adjustment

Fairly strong evidence maintains the relation between social support, beliefs, coping and adjustment. Social support was related to self-efficacy89, 93, 97 and hope,98 but inconsistently to self-esteem (N=77–270).98, 99 Social support was associated to coping in seven studies (N=37–255).71, 75, 81, 83, 99, 100, 101 In one longitudinal study, the perception of social support predicted coping.101 Social support was negatively associated with emotional coping,83, 99 positively with fighting spirit and sense of humor.71, 75 Informational support was related to more problem-oriented coping99 and an internal coping style was connected to higher levels of support compared with external coping.100

Social support was correlated with better adjustment to disability in five studies (N=70–255)47, 64, 75, 77, 102 and mediated the relation between leisure engagement and adjustment.102 Support from friends was associated with acceptance of disability and emotional support with personal growth.75, 77

Functioning, activity and participation

Evidence of the relation between social support and functioning is consistent. In all, 10 cross-sectional studies examined aspects of functioning and integration in relation to social support (N=37–290). Persons who have more social support,69, 102 more reciprocal relationships,103 more support from peers,74 and fewer relationships in which other persons provided more help103 were more likely to be mobile, productive and interested in leisure activities. Satisfaction with social support was associated with functional independence.71, 104 Social and emotional support was linked to better psychological and social functioning.74, 86 Together with self-efficacy and perceived health, social support was linked to and accounted for 25% of the variance in psychological well-being.80, 88 However, these studies address very different aspects of functioning, activity and participation.

Sociodemographic and injury-related characteristics

The existing evidence indicates, not always in a consistent manner, that social support is not related to sociodemographic and lesion-related variables. Results of the relation to gender,76, 77, 86, 87 education,75, 80, 82, 86, 87, 89 race62, 89, 93 or employment status87, 89, 109 were not significant. The relations to marital status and age were inconsistent, depending on the source and type of support.62, 70, 75, 80, 86, 87, 93, 98, 104 Social support was not related to age at injury65, 75, 77, 85, 87, 89 and level or completeness of injury.62, 70, 76, 77, 85, 89, 107

Overall perceived social support was higher in persons with SCI than in controls without SCI.88, 105 People with SCI experienced more support than did people with stroke.104 In three longitudinal studies (N=40–120), overall social support did not change over time.63, 101, 110 However, when source and function of support are differentiated, friend and informational support decreased after injury.84, 101 The family was the most frequently mentioned (N=308)80 and most important source (N=52–100)106, 111 of social support.

Social support intervention

One intervention study was found, which compared a coping effectiveness training with supportive group therapy.61 After treatments, anxiety and depression were reduced. However, no differences between the two treatments were found.

Social skills

Findings about social skills were grouped into four topic domains (Table 7, Figure 3).

Table 7 Summary results on the associations of social skills with aspects of health, functioning and quality of life
Full size table

Mental and physical health

Evidence regarding the relation between health outcomes and specific social skills is strong but not always consistent. Results from five studies (N=51–199) showed a relation between social problem-solving skills and depression or psychosocial disability.112, 113, 114, 115 Higher assertiveness was associated with lower depression.112 However, assertive people were more depressed and psychosocially disabled under conditions of high informational support in a rehabilitation setting.62 Results about social problem-solving skills and the occurrence of pressure sores113, 116, 117 were inconsistent. One longitudinal study (N=188) showed that problem-solving skills were associated with the occurrence of pressure sores in the first 3 years.116 Higher positive problem orientation, a rational problem-solving style, lower impulsive, careless and avoidant style were associated with decreased occurrence of pressure sores.116, 117 Avoidance of problems was associated with urinary tract infections.117

Personal factors

Evidence for the relation between specific social skills and personal factors, such as locus of control or extroversion, is difficult to summarize, because each of the studies investigate different factors. Being assertive was one of the most difficult problems rated by people with SCI (N=35).118 Social problem-solving skills (low negative problem orientation, impulsive, careless and avoidant problem solving, as well as high rational problem solving) were related to acceptance of disability.112, 113, 114 Effective problem solving was associated with assertiveness, confidence and perceived control in problem solving, but not with a person's health locus of control.112, 115 Scoring high in positive problem orientation and using a rational problem-solving style was related to high scores in extroversion, openness to experience, conscientiousness and resilience, low scores in neuroticism, career choice anxiety and generalized indecisiveness.113, 114 Overall, only five cross-sectional studies dealt with the relation between specific social skills and personal factors.

Activity, participation and life satisfaction

Three cross-sectional studies (N=51–206) examined activity, participation and life satisfaction and their relation to social skills.119, 120, 121 Positive problem orientation and rational problem solving were associated with performing more wellness and accident prevention behavior.119 Individuals with SCI who were high in self-monitoring did not differ from those who were low in self-monitoring on free time boredom, but they participated more frequently in recreation activities and socializing, and perceived higher freedom in leisure.120 Communication skills of 158 individuals correlated with life satisfaction, but in regression analysis no significant contribution was found.121

Sociodemographic and injury-related characteristics

The evidence is inconsistent regarding the relation of social skills to sociodemographic and lesion-related variables. In four studies, social problem-solving skills did not correlate with gender and race.113, 115, 116, 119 Assertiveness and effective social problem solving were related to higher levels of education and age.112, 113, 116, 119 People with paraplegia were found to be slightly more effective communicators and problem solvers than were people with tetraplegia,113, 121 and higher the level of SCI, the lower the tendency to act assertively.62, 112

Discussion

This literature review provides a systematic overview on the current state of research in SCI about the relationship of social support and social skills with health and well-being outcomes. The full range of aspects pertaining to social support (that is, type, source, qualifier) and social skills dimensions (that is, topographical, functional, information-processing skills) were addressed in SCI research. However, five times more studies on social support compared with social skills and only one study addressing both concepts in SCI were found. Most studies are of cross-sectional design. In the past 20 years, only few longitudinal and intervention studies were conducted and social support and social skills were mainly measured by self-report questionnaires.

The only study examining both social support and social skills in SCI indicated that the social skills of an individual are correlated with the outcome of social support exchange.62 Research in the general population19, 20 and in mental illness18, 21 shows that social skills are correlated with social support. However, the relationship between social skills and social support and the association of the two with health and quality of life remain unclear. Therefore, a key finding of this review is that there is a need for future research in SCI to confirm whether effective social skills can mobilize social support and how.

In general, the results confirm that social support and social skills are positively related to physical and mental health. Social support seemed to decrease the risk of mortality, facilitate coping and enhance life satisfaction and subjective well-being in people with SCI. Sociodemographic and lesion-related variables were rarely associated with social support but can be related with social skills. Whether a person with effective social skills does mobilize social support and is in turn healthier and more satisfied remains to be explored.

The studies included in this review have fulfilled at least 50% of the STROBE quality criteria. However, most of the studies are of cross-sectional design and hence, do not clarify cause–effect mechanisms. For example, the results show that social support is associated with higher levels of well-being. However, the literature also indicates that well-being of a person with SCI predicts the availability of social support.80 In addition, moderating and mediating effects or change and time effects stay concealed with cross-sectional studies. Longitudinal research is required in future. This could be facilitated, for example, by building registries or research platforms similar to those in the US Model Systems.122 Such platforms should include a comprehensive set of assessment domains addressing all dimensions of functioning and disability, as well as contextual and personal factors, for example, according to the ICF (International Classification of Functioning, Disability and Health).123

The findings of this review in SCI are largely in line with current research. The relationship of social support with mortality,9, 10 physical and mental health,3, 4, 5, 6, 124, 125, 126 coping127, 128, 129 and life satisfaction7 has been confirmed in the general population and in other diseases.130, 131, 132 Social support is also related to the available strengths and internal resources of persons with SCI.13 Social skills have been found to relate to physical and mental health in the general population58, 133, 134 and other diseases.21, 23, 24, 25, 26, 27

There are only few studies addressing social skills in SCI. The concept of social skills is difficult to define. Social skills, social competence, social intelligence or social performance are often used interchangeably.23 In addition, social skills does not seem to change in SCI. However, social skills are important in the development and maintenance of interpersonal relationships, in general and in SCI.121 Social skills training has shown its effectiveness in general population,135 with children and adolescents136 in relation to mental24, 37 and physical health.137, 138, 139 In SCI, there are two publications but small sample sizes.140, 141 Intervention studies in social skills have also reported improvements in social support ratings.142 Therefore, social skills training could be integrated in treatment plans143 and could also prepare people with acquired SCI for difficult social situations.

Only specific social skills, such as social problem solving or assertiveness were examined. General social skills as they are assessed, for example, by the Social Skills Inventory144 are not addressed in SCI. Clinical interviews or behavioral observation (for example, role play) are comprehensive assessments used in general practice to assess social skills.145 They are demanding in terms of administration and analysis. Few instruments exist that have been designed specifically for SCI, such as the SCIQ (Spinal Cord Injury Assertion Questionnaire).29

Looking at issues of measurement, in SCI and in the general population, mostly the different types of social support are measured. However, instruments also assessing the quality and not only the quantity of social support should be considered in research, as quality63, 68, 71, 72, 73 not quantity63, 68, 76, 81 of social support is linked to mental health. Overall, a ‘gold standard’ in assessing social support does not exist; nonetheless, investigators must determine what aspect of social support they consider as important to be evaluated in relation to their specific research question.142

Demographic characteristics of a person do not affect the amount of support perceived. However, there seems to be a difference in levels of importance of social support for older compared with younger people.146 The relation between being married and social support was inconsistent. This may reflect the difference in assessment of quantitative and qualitative aspects of marital support and being married is only a source of social support if the marriage is a good one.147, 148 However, social support, in specific marriage, is also affected by SCI.104 Divorce rates in people with SCI range between 8 and 48%.11 Although SCI represents a major burden to the spouses,149 partners report also some positive changes, such as more open and honest communication.150 To strengthen marriage as an important social support system, comprehensive support should be provided to caregivers, for example, in terms of relationship counseling.

Social support was slightly related to the type of disability (SCI vs stroke), but not to the level of injury. Assistance to people with SCI, due to the physical impairment, is in large part support that is paid. However, this kind of support was not included in this review.

Although social support was consistently found to be positively related to life satisfaction, subjective well-being and quality of life, one study showed somewhat contradictory findings.86 Although social support is considered a positive concept, social relationships may also serve as sources of stress.151, 152 The results show that higher instrumental and informational support and emotional support from friends are related to lower satisfaction with social life.86 This type of social support might lead to unsatisfactory social life, because it may act as a constant reminder of the presence and impact of the disability. It reflects a relative inequality of exchange between the provider and recipient and fosters the feeling of being ‘in dept’.64, 153 In SCI, negative experience of support may lead to dysfunctional coping styles,83 exacerbate acceptance to disability,47 enhance the risk of developing post-traumatic stress disorder81, 83 and is related to numerous health conditions.47 Social skills, such as assertiveness, are helpful in general life situations, but can also have negative effects. In rehabilitation, assertive people with SCI may encounter increased attempts from health professionals to control their behavior, while receiving care and treatment.29 As a result, assertive persons may experience more psychological distress.62

Limitations

The study is subject to several limitations: First, search terms were specific to social support and excluded broader terms such as social integration. Terms referring to participation, which is understood as involvement in a life situation,154 were excluded. Articles about social relationships were only included if the relationship provides support in some way. For example, papers comprising the term ‘marital status’ were only included, if support by marriage, such as spousal support, was examined.

Second, the selection criteria regarding age, qualitative studies and sample size are disputable. Basic social skills are learned and more or less effective for good developmental outcome and adaptation around the age of 13 years.155, 156, 157 However, the development of social skills is not completed at a certain age. Qualitative studies provide detailed insights and the possibility to generate hypothesis158 and could be addressed in a separate review. The decision to solely include studies with N⩾30 is based on reasons of generalizability and power of analyses.159

Third, using STROBE for quality grading can be problematic. STROBE has been applied to assess study quality in a wide variety of systematic literature reviews.160, 161, 162 However, STROBE addresses the reporting of studies rather than their quality.163 A standardized interpretation manual for STROBE scoring does not exist. In addition, results from studies with higher scores in the quality assessment tool were not weighted differently than were results from studies with lower quality.

Conclusions

This systematic literature review provides guidance for future research. The focus should be set on longitudinal and intervention studies. Social skills can mobilize social support,18, 19, 20, 21, 62 which is an important facilitating environmental factor in people with SCI and in the general population.3, 4, 5, 6, 7, 8, 9, 10

People with SCI need all the social support they can get. Therefore, strengthening the social support system of a person, for example, the family, is an important aim in rehabilitation counseling. However, patients should not be seen as passive recipients of support and dependent on others, but rather as persons who can actively and autonomously shape their social relations by using their social skills. Thus, social skills can be an important intervention target in rehabilitation.

However, the relationship between social skills and social support, and how this interrelation operates in relation to health and functioning, have not been fully understood in SCI. The results of further studies could contribute to the development of targeted interventions to enhance functioning, health and quality of life of people with SCI.