Development and change of young adults' preventive health beliefs and behavior: influence from parents and peers.

A longitudinal data set is employed to explore the sources of stability and change in young adults' health beliefs and behavior concerning drinking, diet, exercise, and wearing seat belts. There is substantial change in the performance of health behaviors during the first three years of college, and peers have a strong impact on the magnitude of that change. In total, however, parents are much more important than peers as sources of influence over these beliefs and behaviors. Of the various social influence processes considered, the direct modeling of behavior appears to be the most important avenue of influence for both parents and peers. These data, along with previous papers in our research program, suggest a pattern of gradually increasing parental influence on their children's health beliefs and behavior while the children are living at home, and the persistence of that influence at least through the college years.

A multivariate analysis of adolescents' orientations toward physician use.

Health professionals know surprisingly little about the origins of the health beliefs and orientations they often seek to change, and they do not know about when, or if, changes in these orientations are likely to occur. This study is aimed at increasing our understanding of the origins and development of health beliefs and attitudes by focusing on the impact of parents' orientations toward physician utilization (inclination to consult a doctor when experiencing somatic symptoms) on the comparable orientations of their adolescent children. Hypotheses regarding direct parent-adolescent links are examined within a multivariate structural equations model relating demographics, adolescents' interpersonal competencies, and parental variables to adolescents' orientations toward physician utilization. Evidence for the development of these orientations is obtained by comparing influences across two groups of adolescents: 12- to 13-year-olds and 16- to 18-year-olds. Findings indicate that adolescents' orientations toward physician use are associated with those of their parents and that this link is generally stronger and more specific during late adolescence. The model also suggests that these orientations conform to social and cultural variables related to differences in social stratification.

Further explorations of common-sense representations of common illnesses.

A simple method is presented for measuring people's illness cognitions--their common-sense representations of common illnesses. Data were collected from 1,628 different respondents who described a recent illness form 1 to 3 separate times over a 17-month period. A free-clustering task performed by a set of naive participants confirmed that these cognitions fall into the five components that have been previously noted: identity, time line, consequences, cause, and cure. These five components are found to be reasonably stable over time and across different illness episodes. Several consequences of these illness cognitions, in terms of changes in health-locus-of-control beliefs and different propensities to visit a doctor, are also noted. Specifically, controllable attributions for getting sick and personal responsibility attributions for getting better are associated with increased beliefs in Self-Control Over Health and decreased beliefs in Chance Health Outcomes; people with strong Identity and Cure components in their common-sense representations of common illnesses have a greater propensity to visit a doctor when feeling ill.

Persistence of inaccurate beliefs about the self: perseverance effects in the classroom.

The perseverance of erroneous self-assessments was examined among high school students. Subjects were first exposed to either highly effective or thoroughly useless filmed instruction, leading, respectively, to their consequent success or failure. No-discounting subjects received no assistance in recognizing the relative superiority or inferiority of their instruction. Discounting subjects, by contrast, were subsequently shown the opposite instructional film, highlighting the obvious differences in instructional quality. Subsequent measures revealed that all subjects recognized the effectiveness or ineffectiveness of their instruction, although this contrast was clearer for discounting subjects. Nevertheless, both discounting and no-discounting subjects continued to draw unwarranted inferences--in line with their initial outcomes--about their personal capacities, immediately afterward. Dissociated and disguised measures of academic preferences and perceptions completed weeks later produced even more dramatic results: The continuing impact of initial outcomes was generally greater for discounting than no-discounting subjects.

Health as a value: methodological and theoretical considerations.

The concept of value placed on health is very important in several different theoretical approaches to the study of health behavior. In practice, however, health value is generally assumed to be universally high rather than being directly measured. If this assumption is incorrect, then theories that include health value have rarely been adequately tested. This paper presents a short 4-item Likert scale designed to measure the value placed on health. Norms from the utilization of this scale in five different samples are presented. Health value is found to increase with age among girls, but the increase apparently stops by late adolescence, before full adult levels of health value are achieved. Middle-aged women place a higher value on health than do middle-aged men, although no comparable sex difference appears in a sample of undergraduates. Consistent with theoretical predictions, both health locus of control beliefs and beliefs in the efficacy of certain preventive health behaviors correlate more highly with the performance of those same behaviors 5 to 9 months later among respondents who place a high value of health relative to those who do not value health so highly. However, this interaction is found only when it can be safely assumed that health is the primary value underlying the behavior. The importance of considering a variety of values in addition to health as possible motivators of preventive health behavior is stressed.

Origins of health locus of control beliefs.

A burgeoning literature is discovering the links between locus of control beliefs and a variety of different health beliefs, health behaviors, and health itself. If locus of control beliefs are important to health, it is crucial to understand where those beliefs come from. This study explored the origins of health locus of control (HLC) beliefs, using the Lau-Ware multidimensional HLC battery. A structural equations analysis showed that beliefs in Self-Control Over Health were positively related to early health habits involving self-care and medical professionals and negatively related to prior family experiences with sickness. Beliefs in Provider Control Over Health were positively related to early health habits involving medical professionals and negatively related to prior sickness experiences in one's family. Beliefs in Chance Health Outcomes were positively related to prior sickness experiences, however, supporting a view of HLC beliefs as fairly stable individual-difference measures developed relatively early in life.

Psychosocial problems in chronically ill children: physician concern, parent satisfaction, and the validity of medical records.

This study concerns the psychosocial aspects of treatment for chronically ill children. The English-speaking parents of 44 children 5-13 years of age being seen at five specialty clinics at a large county hospital in Los Angeles, and their attending physicians, were the subjects in this study. The parents were interviewed concerning their expectations for the current visit, and the doctor-patient interaction was tape-recorded. Identical categories of information were abstracted from the tape recording and from a chart review of the patients' medical records. Although parents expected 76% of the psychosocial aspects of care to be covered by the doctor, only one fourth were actually discussed in the visit. These unfulfilled expectations were associated with lower satisfaction with medical care received (r = .47, p less than 0.01). Finally, while doctors recorded about 80% of discussions of symptoms and physical examinations in the patient's medical record, they recorded only 25% of discussion of psychosocial problems.

Refinements in the Measurement of health-specific locus-of-control beliefs.

A health-specific locus-of-control (HLC) questionnaire was constructed to measure beliefs about Self-Control Over Health, Provider Control Over Health, Chance Health Outcomes and General Health Threat. A factor analysis of items confirmed these dimensions, and corresponding scales were shown to be reliable. A factor analysis of HLC scales and other measures was used to evaluate construct validity. Persons believing in provider control over health outcomes tended to have strong favorable attitudes toward the quality of medical care. Chance Health Outcomes was associated with more general chance and controllability measures. General Health Threat seemed to tap susceptibility to disease and beliefs that doctors are unable to diagnose those diseases. People who believe in Self-Control Over Health tended to reject the role of chance and health threat in determining health outcomes and to believe in the efficacy of doctors, although they view them as inaccessible. Results are compared with those of previous studies, and some applications of the questionnaire are discussed.