Perceived control in relation to socioeconomic and behavioral resources for health

https://doi.org/10.1016/S0277-9536(00)00280-XGet rights and content

Abstract

Perceived control is a personality characteristic that contributes to well-being, but few studies have attempted to integrate the functions of perceived control with those of other determinants of health. This research tested two hypotheses about the functions of perceived control: (a) individual differences in perceived control would account for socioeconomic differences in self-rated health status; (b) performance of health-related behaviors would account for the health benefits of perceived control. Using data from adult, nonproxy respondents in the National Population Health Survey of Canada (1995; n=11,110), confirmatory factor analysis supported a measurement model of self-rated health status composed of two correlated factors: physical health (i.e., chronic conditions, restricted activities, self-rated general health, physical functional capacity) and mental health (i.e., distress, depression). Structural equation modeling supported the first hypothesis, but not the second, regarding perceived control as a determinant of self-rated physical and mental health. Health-related behaviors partially mediated age differences in self-rated health, but different behaviors functioned in this way for men than for women. The findings suggest that psychological process, that of perceiving control over life events, underlies social inequality in health. Health-related behaviors appear not to serve as the primary mechanism through which perceived control influences health

Introduction

To have an enduring sense of control over the important events in one's life is a personality characteristic that contributes to well-being. A large body of research spanning several decades consistently supports this association (see Rodin & Salovey, 1989; Rodin, Timko, & Harris, 1985, for reviews). Through this research the construct of perceived control has undergone considerable refinement. Conceptually and empirically, researchers have identified multiple facets of perceived control, offering an increasingly precise grasp of its relationship to positive outcomes in various life domains (Skinner, 1996). As compared with such efforts at refinement, however, few studies have attempted to integrate the functions of perceived control with those of other resources for health.

The goal of the present research is to examine how perceived control functions specifically in relation to sociodemographic and behavioral determinants of health. We are aided in this effort by available data from the National Population Health Survey of Canada (NPHS, 1995). These data, from a national probability sample, include information about respondents’ sociodemographic background, perceived control, and performance of health-related behaviors. Our literature review proposes a theoretical model of relationships among these variables, shown in Fig. 1. Our analysis then establishes the fit of this model to the NPHS data, in predicting multiple dimensions of respondents’ self-rated health status. Throughout, we emphasize two paths that correspond to hypothesized functions of perceived control. The first regards perceived control as a mediating factor in the relationship between socioeconomic status and health. The second regards perceived control as a source of motivation, accounting for individuals’ performance of health-promoting behaviors.

The perception of control ultimately derives from an individual's history of success or failure at attaining goals. Individuals who experience repeated success in goal attainment may develop a generalized expectancy that they can influence the occurrence of positive outcomes or rewarding events in their lives (Rotter, 1966). By the same token, individuals who experience repeated failure in their attempts to secure rewards or to escape punishments may develop a generalized expectancy that they are helpless (Seligman, 1975). These generalized expectancies become important determinants of coping in the face of hardships, including threats to health (Folkman, 1984). When negative events happen, both the stress that people experience (Matthews, 1982) and their persistence in seeking an adaptive response (Lau, 1988) are likely to depend on their prior belief that they are in control of their lives.

The link between perceived control and positive health outcomes is empirically well-established. The design and evaluation of control-enhancing interventions constitutes one major approach to this linkage in previous research. For example, Rodin and Langer (1977) encouraged senior nursing home residents to take responsibility for daily life activities, such as caring for plants. Seniors who participated in this intervention showed sustained benefits, compared to a control group, not only in physician-rated health status, but also in mortality rates over an 18-month period. Based on this research, Searle, Mahon, Iso-Ahola, Sdrolias, and van Dyck (1995) developed a control-enhancing intervention for older adults who were living independently. These individuals had indicated on a prior survey that they were experiencing a problem or had already ceased with participating in a favorite leisure activity; thus, they were at risk for losing an important component of their independence. The intervention was designed to foster a sense of independence and life satisfaction by helping participants to plan effectively for future recreation. Compared to participants in a random control group, those who received the intervention showed significant increases in leisure-related perceived control and general life satisfaction. Thus, control-enhancing interventions may contribute to recipients’ quality of life, as well as longevity.

The use of individual difference measures represents a second major approach in research linking perceptions of control to health. Indicators of both mental (Weary, Marsh, Gleicher, & Edwards, 1993) and physical (Helgeson, 1992) health have been shown to correlate positively with individual difference measures of perceived control. Such correlations have been observed across select populations, including the elderly (Baltes & Baltes, 1990), as well as across specific diseases, for example arthritis (Tennen, Affleck, Urrows, Higgins, & Mendola, 1992) and coronary heart disease (Taylor, Helgeson, Reed, & Skokan, 1991). A recent study by Chipperfield (1993) found that elderly individuals’ lack of perceived control was associated prospectively with mortality from all causes over a 12-year period. Kaplan, Ries, Prewitt, and Eakin (1994) further demonstrated, among patients with chronic pulmonary obstruction disease, that self-efficacy compared favorably with physiological measures of pulmonary function as a predictor of survival over a 5-year period.

This body of research clearly establishes a positive relationship between the perception of control and health status, which holds across populations whose health is at risk. Less clear, however, is the extent to which the same relationship holds for the predominantly healthy general population. As well, previous research documenting the influence of perceived control on health outcomes has remained surprisingly separate from research identifying the pathways through which this influence is thought to occur. The separation between outcomes and processes in previous research has made it difficult to determine the importance of a variety of potential mediators associated with perceived control. The present research attempts to bridge these gaps in two ways. First, by examining perceived control and self-rated health status in a broad sample, the present research tests the degree to which sociodemographic determinants of population health are mediated by individual differences in this coping resource. Second, by evaluating the relationship of perceived control to adaptive behaviors and self-rated health status in a single study, the present research tests the degree to which individuals’ performance of healthy lifestyle behaviors may account for the health benefits of perceived control.

Much research attests to the fact that health problems are unevenly distributed in North American society. Lower socioeconomic status and older age are reliably associated with poorer self-rated health, while the presence of sex differences in health status varies across indicators (see Roberge, Berthelot, & Wolfson, 1995, for a recent analysis of Canadian data). Pearlin and Schooler (1978) further examined the distribution of multiple coping resources in the general population. These investigators demonstrated that people with high socioeconomic status enjoyed greater access to feelings of mastery than did people with low socioeconomic status. Older age and female sex were also associated with lower mastery in this research, although these correlations were small. Subsequent studies have reported similar patterns of sociodemographic variation in perceived control (Rodin et al., 1985).

The question raised by these findings is whether sociodemographic risk factors predispose individuals to poorer health outcomes indirectly by means of reduced psychological coping resources such as perceived control. Based on recent reviews in the field of medical sociology, this hypothesis appears especially promising in regard to low socioeconomic status as a risk factor. Evans and Stoddart (1990) observe that the extent of inequalities in health across the socioeconomic classes has remained stable over a long period of time, even though the absolute level of income and principal causes of death have changed. These authors conclude that factors underlying the relationship between socioeconomic status and health must influence susceptibility to a broad range of diseases, and must be more related to social position than to absolute resources for accessing health care. Some previous research indicates that control beliefs may fulfill this mediating role. For example, Grembowski, Patrick, Diehr, Durham, Beresford, Kay, and Hecht (1993) reported that elderly individuals’ ratings of their perceived ability and likelihood to control health-related behaviors mediated the relationship between their socioeconomic status and functional health.

Socioeconomic status differs from age and sex in several respects that may account for the weaker relationship of the latter to perceived control. Pearlin and Schooler (1978) note that wealth and education are achieved statuses, providing implicit evidence of control, whereas age and sex are ascribed. A more complex argument derives from Heckhausen and Schulz's (1995) theory that experiences of limitation, which threaten one's sense of primary control over the environment, may normally be compensated through the use of secondary control strategies. These strategies are inner-directed efforts at changing oneself to fit one's circumstances — for example, through setting attainable goals or finding the “silver lining” in every cloud. People thus can maintain perceived control on an even keel by increasing their reliance on secondary strategies as they encounter constraints due to age or sex. When a loss in primary control is due to socioeconomic disadvantage, however, it may be especially difficult to use secondary strategies such as downward social comparison as a way of accommodating the loss. Thus, in the present analysis, we hypothesized that perceived control would account for the relationship between socioeconomic status and multiple aspects of self-rated health in the general population.

A number of routine behaviors have been shown to influence people's self-rated health (Belloc & Breslow, 1972; Segovia, Bertlett, & Edwards, 1989). Indeed, qualitative research by Krause and Jay (1994) suggests that among younger individuals, performance of health-related behaviors is a primary basis for self-evaluations of health. Four such behaviors were measured in the NPHS (1995): respondents’ frequency of leisure physical activity, smoking, consuming alcohol, and having their blood pressure examined by a health professional. Comparable behaviors have been linked to control constructs in past research (Ziff, Conrad, & Lachman, 1995).

The rationale for viewing such behaviors as a mechanism of the relationship between perceived control and health derives additional theoretical support from research in developmental and social psychology. As Dweck and Leggett (1988) have shown in the context of intellectual achievement, children with a mastery orientation take satisfaction in using active and adaptive strategies to solve challenging problems. Yet children with a helpless orientation, despite their equal intellectual ability, adopt the frustrating course of repeating strategies that have already failed to solve these problems. Extending these patterns into adulthood, it seems plausible that individuals who are high in perceived control may be especially likely to use adaptive behavioral strategies on a routine basis to maintain or enhance health. The self-reported behavioral data in the NPHS (1995) offer an opportunity to provide confirming or disconfirming evidence for this pathway.

Social psychological research has further established relevant limiting conditions on the relationship between perceived control and health. Studies of medical patients, for example, suggest that the adaptiveness of beliefs in personal control will be greatest when they are realistic with respect to actual levels of control in treatment (Helgeson, 1992; Thompson, 1993). Exaggerated perceptions of control, in contrast to realistic ones, are likely to aggravate the stress felt by patients whose efforts at recovery fail to produce positive results. The importance of actual control seems again to implicate behavioral responses, such as taking action to prevent or remedy health problems, in accounting for the relationship between perceived control and health status.

The question raised by this research is whether perceived control also serves to motivate healthy lifestyle behaviors among people who are not facing an immediate health threat. A series of path analyses reported by Goldsmith Cwikel, Dielman, Kirscht, and Israel (1988) provides limited empirical support for this hypothesis. The path analyses compared the direct and indirect effects of three belief statements concerning perceived control on self-rated general health. The indirect model situated an index of positive health practices and an index of preventive health behaviors between the perceived control and health status measures. Among female respondents, though not among males, the indirect model accounted for greater variance in respondents’ self-rated general health than did the direct model. In the present analysis, we sought to provide more compelling evidence either for or against the hypothesis that individuals’ performance of health-promoting behaviors would account for the health benefits of perceived control.

As a final step in our theoretical model, we consider sociodemographic influences on health-related behavior. This path is important for practical purposes of targeting health promotion programs to segments of the population who are at risk, for example, regarding smoking and lack of physical activity as leading preventable causes of mortality in North America (Centers for Disease Control and Prevention, 1997; US Department of Health and Human Services, 1996). Inclusion of this path permitted an exploratory comparison of perceived control and health-related behaviors as independent channels through which sociodemographic factors may influence self-rated health status.

Section snippets

NPHS sample and procedure

The target population of the NPHS (1995) was household residents in all provinces, excluding residents of Indian Reservations, Canadian Forces bases, the Northwest or Yukon Territories, or long-term care institutions. A multi-stage probability sampling method was used to select dwelling units at random within a geographically stratified sample of census areas. After an initial household survey (response rate=88.7%), individuals over the age of 12 were selected randomly within households to

Results

Table 1, Table 2 present descriptive statistics and zero-order correlations of all model variables for male and female respondents, respectively. Table 3 presents factor loadings of the six self-rated health status measures on physical and mental health.

Discussion

The motivation for this investigation was to understand the functions of perceived control as a personal resource for maintaining and enhancing health. How successfully, and under what conditions, do individual differences in the perception of control account for variations in population health? Through what intervening effects on individuals’ decision-making and behavior does the perception of control ultimately benefit health? The results of this analysis offer insights and a basis for

Conclusion

The results of the present analysis have both theoretical and practical implications. These results provide evidence that a psychological process, the perception of control over life events, may underlie socioeconomic differences in health. Perceived control fits the prescription offered by sociological analyses of class differences in health, as a factor that influences susceptibility to a broad range of diseases, and that varies as a function of individuals’ relative social position rather

Acknowledgements

This research was supported by a Health Canada, National Health Research and Development Program grant No. 6607-1781-NPHS to the first four authors. The authors are indebted to Jacquie Vorauer for comments on an earlier draft of this manuscript, and to Rod Clifton for statistical consultation.

References (55)

  • N.B Belloc et al.

    Relationship of physical health status and health practices

    Preventive Medicine

    (1972)
  • R Evans et al.

    Producing health, consuming health care

    Social Science and Medicine

    (1990)
  • American Psychiatric Association. (1987). Diagnostic and statistical manual of mental disorders (3rd rev. ed.)....
  • J.L Arbuckle

    AMOS 4.0 programming reference guide

    (1995)
  • Baltes, P.B., & Baltes, M.M. (1990). Psychological perspectives on successful aging: The model of selective...
  • R.M Baron et al.

    The moderator-mediator variable distinction in social psychological researchConceptual, strategic, and statistical considerations

    Journal of Personality and Social Psychology

    (1986)
  • M.P Beaudet

    Depression

    Health Reports

    (1996)
  • Berthelot, J., Roberge, R., & Wolfson, M.C. (1993). The calculation of health-adjusted life expectancy for a Canadian...
  • Centers for Disease Control and Prevention. (1997). Facts about cigarette mortality. Atlanta, GA: CDC Media Relations...
  • J.G Chipperfield

    Perceived barriers in coping with health problemsA 12-year longitudinal study of survival among elders

    Journal of Aging and Health

    (1993)
  • T.E Duncan et al.

    Serial fluctuations in exercise behavior as a function of social support and efficacy cognitions

    Journal of Applied Social Psychology

    (1993)
  • C.S Dweck et al.

    A social-cognitive approach to motivation and personality

    Psychological Review

    (1988)
  • S Folkman

    Personal control and stress and coping processesA theoretical analysis

    Journal of Personality and Social Psychology

    (1984)
  • J.M Goldsmith Cwikel et al.

    Mechanisms of psychosocial effects on healthThe role of social integration, coping style and health behavior

    Health Education Quarterly

    (1988)
  • D Grembowski et al.

    Self-efficacy and health behavior among older adults

    Journal of Health and Social Behavior

    (1993)
  • Grob, A., Little, T.D., Wanner, B., Wearing, A.J., & Euronet (1996). Adolescents’ well-being and perceived control...
  • J Heckhausen et al.

    A life-span theory of control

    Psychological Review

    (1995)
  • V.S Helgeson

    Moderators of the relation between perceived control and adjustment to chronic illness

    Journal of Personality and Social Psychology

    (1992)
  • E.L Idler et al.

    Self-rated health and mortalityA review of twenty-seven community studies

    Journal of Health and Social Behavior

    (1997)
  • K.G Jöreskog et al.

    LISREL 7A guide to the program and applications

    (1989)
  • R.M Kaplan et al.

    Self-efficacy expectations predict survival for patients with chronic pulmonary obstruction disease

    Health Psychology

    (1994)
  • R.C Kessler et al.

    Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the United StatesResults from the National Comorbidity Survey

    Archives of General Psychiatry

    (1994)
  • N.M Krause et al.

    What do global self-rated health items measure?

    Medical Care

    (1994)
  • M.E Lachman et al.

    The sense of control as a moderator of social class differences in health and well-being

    Journal of Personality and Social Psychology

    (1998)
  • R.R Lau

    Beliefs about control and health behavior

  • Long, J.S. (1983). Covariance structure models: An introduction to LISREL. Sage University Paper series on Quantitative...
  • K.A Matthews

    Psychological perspectives on the type A behavior pattern

    Psychological bulletin

    (1982)
  • Cited by (120)

    • The association between childhood socioeconomic status and adult health behavior: The role of locus of control

      2021, Social Science Research
      Citation Excerpt :

      First, adult SES and LOC may interact over the life course (Whitehead et al., 2016) (path F). On the one hand, an external LOC may reflect an actual lack of control caused by a lower SES (Bailis et al., 2001; Murray et al., 2012; Whitehead et al., 2016). On the other hand, a more external LOC may induce a passive and ineffective coping style by not exploiting ones full potential, which may lead to poorer socioeconomic outcomes in life (Cobb-Clark, 2015; Infurna et al., 2016; Schnitzlein and Stephani, 2016).

    • The impact of internal locus of control on healthcare utilization, expenditures, and health status across older adult income levels

      2020, Geriatric Nursing
      Citation Excerpt :

      Those with chance LOC have been characterized in other studies as less likely to utilize preventive services and to opt-out of procedures with significant cost sharing.9,12,21,27 As expected, the associations of internal LOC with better self-rated health and fewer physical limitations were consistently strong across all income groups and in agreement with conclusions suggested from the objective healthcare data.5,7,13,15,17,18,21,26,27,29 Interventions for patients with pain conditions that may promote higher internal LOC, self-efficacy and/or self-confidence have included multidimensional pain management programs that include elements of analgesic drug management, problem-solving and coping strategies, cognitive behavioral therapy (CBT), mindfulness/relaxation, physical therapy or exercise.47–50

    View all citing articles on Scopus
    View full text