In chronic illness, the clients' beliefs in their ability to perform the required health-maintaining behaviors, or self-efficacy; may not be related to their beliefs that health outcomes are contingent on such behavior (response efficacy). Community health nurses (CHNs) may influence health outcome beliefs by substituting the behavior or becoming an external element of disease control. Community nursing services are the major providers of care for older people with leg ulcers (Callam, Ruckley, Harper, & Dale, 1985). Nurses perform activities that clients are unable to perform, such as dressing a wound or bandaging a limb. Self-efficacy beliefs in elders suffering from chronic leg ulcers may not be related to health outcome beliefs or healing, when influenced by CHNs' interactions.
Elderly people comprise 85% of all leg ulcer sufferers (Nelzen, Bergqvist, Lindhagen, & Hallbook, 1991). Venous, arterial or concomitant venous and arterial disease are the most common causes (Baker, Stacey, Singh, Hoskin, & Thompson, 1992). Advances in the management of venous ulcers have resulted in the advocation of nursing interventions, such as limb resting, elevation, and compression bandaging (Blair, Wright. Backhouse, Riddle, & McCoIlum, 1988; Bourne, 1986). Blair et al. (1988) suggest an ankle/brachial pressure index of 0.80, representing mild arterial disease, as the lower limit for therapeutic compression.
The widespread acceptance of the principles of moist wound healing and occlusive dressings has led to these advances being applied to ulcer wound management (Winter, 1971). The nurse usually becomes involved in a lengthy therapeutic relationship. The potential effects of this interaction upon perceptions of control, selfefficacy, and direct and indirect health outcomes will be explored in this study of elderly people with leg ulcers (Figure).
OVERVIEW OF THE UTERATURE
Many authors who have applied the concepts from social cognitive theory, protection motivation theory, the Health Belief Model, and social learning theory, have suggested that health beliefs and control may influence health-maintaining behaviors, and illness /health outcomes (Harrison, Mullen, & Green, 1992; Janz & Becker, 1984; O'Leary, 1982; Rogers, 1975; Rosenstock, 1991; Rosenstock, Strecher, & Becker, 1988; Rotter, 1954; Strecher, DeViIHs, Becker, & Rosenstock, 1986; Strickland, 1978; Strickland, 1989; Wallston, Wallston, Smith, & Dobbins, 1987; Wurtele & Maddux, 1987).
Beliefs about the ability to perform the required behavior (self-efficacy) are associated with beliefs about a positive health outcome being contingent upon behavior performance (response efficacy). The impact that health beliefs have on health outcomes and leg ulcer healing has been demonstrated (Cox, 1986; Johnson, in press). Clients' self-efficacy or response efficacy beliefs were not found to be significant predictors of leg ulcer healing in a comparison of physiological (lipodermatosclerotic changes, edema, wound characteristics) and therapeutic determinants (limb position, compression, dressings, diuretic usage) in older subjects (Johnson, in press).
In elderly people, the impact that interactional and indirect aspects of health beliefs has on health outcomes has been demonstrated (Cox, 1986). The Interaction Model of Client Health Behavior by Cox was applied to a study of community-living elders. Cox suggested that without a positive sense of well-being, personal efficacy, and health behavior control may be difficult to accomplish. Ward's study (1993) into the interactions between the elderly and their professional caregivers demonstrated that the home health care group (n = 14) reported more control in selfcare interactions than the nursing home group (n = 16); more independent self-care behavior was observed in the high self-efficacy group; and the subjects related the lack of use of their abilities, referred to as "nonuse of competence," to the staff's behavior. In contrast, the staff associated this "nonuse" to elder's characteristics.
Brown (1992) analyzed nurseclient interactions and promoted the tailoring of interactions to obtain "congruence between patient characteristics and nurses' actions." The sharing of similar and realistic perceprions of the client's abilities, by nurses and clients, may have implications for long-term, health-behavior maintenance and the appropriate utilization of elders' abilities in leg ulcer care.
Figure. Interactional relationships model
Both aging and chronic disease have been associated with alterations in self-efficacy belief (Lachman, 1986; Marshall, 1991; Wallston et al., 1987). Wallston et al. (1987) proposed that self-efficacy beliefs may not mirror behavior in the presence of an external locus of control. Locus of control is the degree to which individuals perceive events in their lives as being a consequence of their own actions, and controllable (internal control), or as being unrelated to their own behavior and, therefore, beyond personal control (external control - fate, chance, or powerful others) (Lau & Ware, 1981; Rotter, 1954; Strickland, 1978). External health locus of control (powerful other), or beliefs that health outcomes are contingent on other people's behavior, were major predictors of behavioral outcomes in research in older people and chronic disease (Lachman, 1986; Nagy & Wolfe, 1984).
External control is proposed as a positive experience for those struggling with the care of their chronic illnesses (Reid, 1984). Reid (1984) found elderly people in poor health progressively increase in externality over a 5-year period. The increasing acceptance of the competence of others to care for their health was proposed as a possible explanatory mechanism. Brown believed this to be an important conceptual understanding in nurse-client interactions and was labeled "shared control of health processes" (1992). Weisz, Rothbaum, and Blackburn supported the purposeful nature of relinquishing control, suggesting personal power enhancement as the likely outcome (1984). Health professional (powerful other) interaction is proposed to have beneficial effects upon health beliefs and health outcomes in the elderly with chronic disease.
In the clinical situation, the clients rely on the nurse for assistance with their leg ulcer care, perhaps shifting toward shared control or an external health locus of control. The interaction between self-efficacy, beliefs, powerful other, health locus of control, and response efficacy beliefs may influence the healing outcome. Possibly, as long as the client believes in the efficacious nature of the nurse as a powerful other or the self, then the desired outcome is perceived as possible and may be achieved (Figure).
The desired outcome in illness is compliance with sick-role behavior. Gordis (1979) recommends direct and indirect methods of evaluation when determining compliance. The rate of healing could be perceived as a direct outcome measure of compliance with regimes, or the desired outcome. A poor result in direct outcome methods, such as healing rate, may not necessarily be interpreted as noncompliance (Gordis). In this study, beliefe that performing the required health behaviors will result in healing (response efficacy) - an indirect method - also are proposed as a desired health outcome.
This study explored the relationships between clients' and CHNs' efficacy beliefs. An attempt was made to determine the association between self-efficacy beliefs, response efficacy beliefs, and leg ulcer healing. The differences in response to efficacy beliefs and leg ulcer healing rate in clients with varying self-efficacy beliefs and varying powerful other externality also were explored.
Hypothesis 1: Clients' self-efficacy and response efficacy beliefs will be related to CHNs' beliefs about clients' self-efficacy and response efficacy.
Hypothesis 2: Clients' self-efficacy beliefs will be related to the healing outcome and response efficacy beliefs.
Hypothesis 3: Clients with high self-efficacy and low powerful other (health professional) control will not differ for response efficacy beliefs or leg ulcer healing compared with clients with low self-efficacy and high powerful other control.
A comparative, descriptive design was used to compare selected health beliefs and health outcomes in a sample of community-residing clients with leg ulcers.
Sample and setting
A convenience sample of 170 clients (« = 170), aged 60 years or older with leg ulcers was obtained from home visiting nursing services in two major Australian cities. Clients with leg ulcers were asked to participate in the study if they fulfilled the following criteria: no history of sickle cell or thalessemia diseases; no use of oral steroids; ability to understand English or have interpreter assistance available; cognitively able to give consent or have a significant other available to give consent; over age 59; and the presence of a leg ulcer for more than 6 weeks. These subjects, together with their associated visiting nurses (hereafter referred to as CHNs) (n = 120), were studied.
Subject confidentiality and anonymity were assured throughout this study. Informed written consent was sought from all subjects or significant others where appropriate. Assessment of the subjects cognitive function was performed and moderate to severe impairment, when found, resulted in appropriate discontinuation of the interview. Ethical approval for this research had been sought and obtained from the nursing services and institutional ethics committees.
The self-efficacy scale was developed to determine the clients' beliefs about their ability to perform the required health behaviors to heal their ulcer. This short scale, derived from similar studies using this concept, included two therapy-related items on a seven point Likert-type scale (ranging from 1 strongly disagree to 7 strongly agree) - "I have no difficulty in attending to my ulcer dressing when required"; "I have no difficulty in applying my stockings /bandages as suggested." Internal consistency was determined resulting in a standardized alpha coefficient of 0.75 (Johnson, 1993). These items were altered slightly for inclusion in the nurses' questionnaire - "This person feels she/he can attend to her/his ulcer dressing when required" (alpha = .61) (Johnson, 1993). The small number of items was necessary to limit the home visit to 45 minutes.
To determine the evidence of powerful other externality, the Multidimensional Health Locus of Control (MHLC) Scale Subscale of Powerful Others (PHLC) Form B was used (Wallston & Wallston, 1981; Wallston, Wallston, & DeVìffls, 1978). A high score in this scale determines that health-related behaviors are under the control of powerful others. This scale uses a Likert-type scale ranging from 1 strongly disagree to 6 strongly agree. Alpha reliability coefficients found in various studies range from 0.69 to 0.73 (Wallston & Wallston, 1981).
The client's perception of response efficacy was determined by two items based on interventions - "By resting my legs, my ulcer is likely to heal," and "By regularly dressing my wound, my leg ulcer is likely to heal." As before, the items used a seven-point Likert-type scale (ranging from 1 strongly disagree to 7 strongly agree), (alpha =0.61) (Johnson, 1993). Similar items with slight alterations were included in the nurses' questionnaire - "By regularly dressing the client's wound, her or his leg ulcer is likely to heal" - (alpha = 0.54) (Johnson, 1993).
Docf or-Dicuynosecf Health Problems In Sample
To determine the accuracy of the information being gained and to confirm that informed consent was being given, the level of cognitive function of me client was determined by the Short Portable Mental Statua Questionnaire (SPMSQ) (Pfeiffer, 1975). Concurrent validity has been demonstrated by comparing the scale to the client diagnosis of organic brain syndrome. There was 92% agreement between the SPMSQ score and the clinical diagnosis of definite impairment and 82% agreement when the SPMSQ indicated no or mild impairment (Pfeiffer, 1975). Test-retest correlations of 0.82 and 0.83 have been demonstrated (Pfeiffer).
Healing was measured by the technique of stereophotogrommetry (computerized topographical mapping). Stereophotogrornmetry has been demonstrated to be an accurate method of area measurement (0.34% error compared to known area model ulcers) and volume measurement (5% error compared with model ulcers and fluid displacement) (Bulstrode, Goode, & Scott, 1987; Johnson & Miller, in press). Healing was defined as a 10% reduction in the surface area over the study period.
To assess the health status of the clients, a single item determining health perceptions (excellent to poor) was included. This item has been used extensively in research in the elderly and has been found to be related to disability level and number of illnesses (Ferraro, 1980). The client also was asked about current doctor-diagnosed illnesses.
Ankle /brachial pressure indices were determined using a Medasonics Ultrasound Stethoscope Doppler Blood Flow Detector. This technique requires the torso to be at a 40° angle to the horizontal limbs. The occluding systolic blood pressure was detected using a sphygmomanometer and stethoscope over the brachial artery. The most audible systolic occluding pressure of the dorsalis pedís, posterior tibial, or the peroneal vessels was determined. The ratio of the ankle to brachial pressure of <0.90 has been associated with arterial disease (Johnson & Patten, 1977). Concurrent validity has been demonstrated with TcPO2 measurement (Alien & Goldman, 1987). Stability of doppler measurement previously has been established (Rooke & Osmundson, 1989).
Distribution of Ulcer Sizes in the Sample
Clients were visited on two occasions (Week 1 and Week 4 or 5). If the client had more than one ulcer, an ulcer site was randomly selected. A box containing defined ulcer positions - most proximal ulcer right leg, most distal ulcer left leg - was used.
On the first visit, dopplers were performed, dressings were removed, and stereophotographs of the wound taken. After the completion of the dressing by the CHN, the investigator used a structured interview guide to finish the data. The nurse's questionnaire was completed by the nurse and returned to a collection box located at the health center. On the second visit only stereophotographs were taken. The overall time period for data collection was 7 months. If possible, the exact time of day and day of the week was kept for the 4-week follow-up visit. In a small number of cases, follow-up measurements were ±48 hours from the planned scheduled.
All analyses were performed using SPSS (Statistical Package for the Social Sciences). Descriptive statistics, Pearson Product Moment Correlation Coefficients, chi-square procedures, and independent / tests were used in this study,
Healing rate per 28 days was calculated with the following formula: (ulcer area at follow-up initial ulcer area) X 28 divided by initial ulcer area X days between measurements.
This calculation results in a negative score for subjects who are healing and a positive score for subjects not healing. For example, if the ulcer surface area was less in Week 2 than Week 1, a negative difference would be found in the healing rate.
Sample and setring
Clients. From the sample of 170 subjects who agreed to participate in this study, the mean age was 78 (SD = 8 years), the range was 60 to 96, and 58% of the participants were female. Subject losses, due to unexpected events, such as hospitalization, totaled 8% (n = 14).
Fifty-nine percent of the subjects rated their health as good, very good, or excellent. Health problems found in the subjects are listed in Table 1. Ten percent of clients had rheumatoid arthritis, 14% had diabetes mellitus, and 26% had atherosclerosis or hardening of leg arteries. Osteoarthritis (55%) and hypertension (41%) were the most frequently occurring health problems. The mean ankle/ brachial ratio was 0.84 with a standard deviation of 0.29. Fifty-one percent of the subjects had more than one ulcer (range 1 to 12 total ulcers). The duration of ulcer in months ranged from 1 to 384 months. Ninetyfive percent of subjects were found to have intact cognition or mild impairment determined by the SPMSQ.
Ulcer sizes varied from 3 mm2 to 5367 mm2 with a mean of 636.7 mm2 and a standard deviation of 868.41 mm2. Table 2 demonstrates the distribution of ulcer sizes for the sample.
Nurses. From the sample of 120 CHNs, 48% had less than 3 years experience as a visiting nurse; 69% of nurses had 3 years' hospital training in nursing; the remaining nurses having a diploma, degree or master's degree. These nurses visited 61% of clients at least three times each week, with the remaining clients being visited once or twice each week; 71% were visited only to treat the ulcer.
INFRACTION OF BEUEFS AND CONTROL
Hypothesis 1 refers to the relationship between clients' and nurses' selfefficacy and response efficacy beliefs. The correlation between client and nurse perceptions of the client's selfefficacy beliefs was r = 0.4710 (p-COOl, p = 118) accounting for 22% of the variance between scores. There was a weak positive relationship between the clients' beliefs about response efficacy and the nurses' beliefs regarding response efficacy (r = .12,p = .19, p = 118), although this did not reach .05 level of significance.
Hypothesis 2, exploring the relationship between client self-efficacy scores and healing status - healers (10% decrease in area or volume) and non-healers (<10% change) - was not supported (X2 (1, n = 140)=3.33, p = . 06). However, changes in client self-efficacy were found to be related to changes in response efficacy beliefs (X2 (1, p = 143) = 4.97, ? = .02). Sixty-eight percent of the low-efficacy group had high response efficacy beliefs compared to 31.6% who had low belief scores. Fifty percent of the high efficacy group had high and 50% had low response efficacy scores. Further analysis demonstrated that changes in self-efficacy beliefs, in the presence of high response efficacy beliefs, were related to healing outcomes (X2 (1, n = 62) = 4.11, p = .04).
The median score for the FHLC was 23 (range 9 to 36), the clients' self-efficacy scale was 6 (range 2 to 12), and the clients' response efficacy scale was 10 (range 2 to 14). All subjects were placed in either Group 1. high powerful others (scores 5*23) and low self-efficacy (scores =£7), or Group 2, low powerful others (scores =£22) and high self-efficacy (scores ^ 8).
Hypothesis 3 proposed that no difference would exist between client response efficacy scores and ulcer healing rates of Group 1 and Group 2. Mean response efficacy scores for Group 1 (M = 10.85, « = 42) were greater than Group 2 (M = 9.20, n = 29) (f = 2.7 l, p = .009) (Table 3). The mean ulcer healing rates of the groups did not differ significantly (Group 1 (M = -.25, n = 38) and Group 2 (M -.32, n = 28) (i = .29, p = .775), resulting in this hypothesis being only partially supported (Table 3).
Group Comparisons of Healing Rafe and Response Efficacy Beliefs
Nurses' and clients' beliefs about clients' self-efficacy behaviors may differ in the elderly. In people with chronic illness, such as leg ulcération, health professionals' influence may alter the relationship between beliefs about the ability to perform required health activities, and beliefs that these activities will result in the desired health outcome. The interactional and indirect effects of control on the clients' beliefs about their own abilities (self-efficacy), and the effect of these on subsequent health outcomes (response efficacy beliefs and leg ulcer healing), have been investigated. This study forms part of a larger study into model development in leg ulcer healing (Johnson, in press).
The relationship between client and nurse self-efficacy beliefs was supported. This result contrasts with the findings of Wahl (1991) in a nursing home and community-residing sample of older people. Congruence between client and nurse perceptions of the client's self -efficacy beliefs support Brown's discussions (1992). From this study, CHNs do consider their client's capabilities when intervening in leg ulcer care.
The low positive relationship between client and nurse response efficacy beliefs (not statistically significant) was consistent with the experience of the investigator. Confusion as to which interventions would result in healing was apparent from both the nurse's and the client's perspective. Also, difficulties in tolerating bandage compression were related anecdotally to beliefs about therapy ineffectiveness.
Research into the mismatch between client and nurse response efficacy beliefs using instruments with improved internal consistency is required. In practice, nurses must develop and use strategies that clarify the effectiveness of specific behaviors, so that both the CHN and the client hold similar beliefs about healing for specific behaviors, such as the use of diuretics in reducing limb edema. Similarly, the use of models or diagrams that demonstrate improvement in venous return using bandage compression and limb elevation may be helpful. Nurses also must explore the link between therapy acceptability and response efficacy beliefs. "Does the discomfort associated with bandage compression result in a perception that the therapy must be detrimental to the healing wound?"
Self-efficacy beliefs were not directly related to healing outcomes in clients with leg ulcers. Changes in clients' beliefs about their abilities to dress a wound or bandage a limb did vary with changes in beliefs that performing mese activities would result in healing; the low ability (selfefficacy) group believed that healing would occur. Also, the low ability group who believed mat dressing the wound and resting their limbs would result in healing, did result in healing. This indirect relationship between self-efficacy beliefs, response efficacy, and healing, requires further investigation in a larger sample capable of stratification on these critical features. A plausible explanation is mat when the CHN is performing the required behaviors to heal the ulcer - dressing the wound or bandaging the limb - beliefs that healing will occur as a consequence of these activities are increased in clients with minimal or no skills in these activities. The assessment of the clients' beliefe about their abilities to perform these functions should be part of the initial visit, thus appropriately using whatever skills do exist and avoiding "nonuse" of skills.
Differences in response efficacy beliefs were found. Those clients who perceived substantial health professional influence, combined with beliefs of limited personal abilities, also believed that healing was likely. CHNs' interactions appear to enhance beliefs in the healing outcome in the presence of low efficacy beliefs in clients. The positive effects of health professional's interventions on health beliefs support the work by Reid (1984) and Weisz et al. (1984). For some clients, relinquishing control of the required sick-role behaviors to the nurse is positive. In other clients, maintaining control over behaviors is important.
No difference was noted in the healing rate of low self-efficacy clients with perceptions that health professionals substantially influenced their health, compared with those who had high self-efficacy beliefs and did not perceive such influence. The likelihood of an ulcer healing, when the client perceived mat he or she was unable to manage the dressing or limb bandaging did not differ when those clients also perceived substantial influence from health professionals and their own limited abilities.
This indirect relationship suggests that self -efficacy does not mirror healing outcomes in the presence of external health locus of control, supporting the notions of Wallston et al. (1987). This understanding also highlights the applicability of Brown and Cox's concepts of shared control of health care processes (Brown, 1992; Cox, 1986). Determining health locus of control in clients with leg ulcers may allow the CHN to ascertain the need for increased or shared control of health processes. It is likely that healing still can occur in clients who recognize the limitations of their own abilities and the potential for nurses to substitute or enhance their existing abilities.
This sample reflected a chronic illness experience and powerful other influence; median score for the PHLC scale was 23, compared with the reported mean score for chronically ill subjects of 22.54 (Wallston & Wallston, 1981). Understanding the role the nurse has as a control element capable of improving illness outcomes requires replication in a similar sample, and study in different client groups. Although nurses are likely to be the most frequent powerful other health professional to interact with these clients, a nursespecific powerful other's scale may find differing results. This present study emphasizes the need for consideration of second order interactions (self-efficacy X powerful other health locus of control x response efficacy) when applying health beliefs to health outcomes in older people with chronic illness.
Several limitations of this study should be noted. The impact of other health beliefe, such as perceived severity and barriers to performing behaviors, has not been addressed in this study. These results must be interpreted in the light of modest alpha coefficients for response efficacy beliefs. These findings do not imply causality, but suggest these concepts are related. Alternative explanations related to confounding factors not assessed in this study also may be plausible. Similarly, maturation of the subjects must be considered. A controlled trial accounting for possible confounding factors may further clarify these issues.
As Cox (1986) and Wahl (1991) suggest, interventions that increase the sense of self-efficacy or competence of the client may have positive health outcomes. The sharing of control over health processes can have benefits for older leg ulcer sufferers who have an awareness of their own limited abilities. Strategies aimed at enhancing beliefs in nurses' abilities may result in improved response efficacy beliefs. Also, client-nurse interactions that correctly parallel beUefs in response efficacy may improve compliance with self-care behaviors in clients with leg ulcers. Any barriers to compliance with these behaviors require investigation, particularly in the area of compression bandaging. This study suggests that beliefs in the efficacious nature of the self, or the powerful other (nurse), can promote beliefs in the desired health outcomes and may indirectly achieve the desired health outcomes.
- Auen, P.I., & Goldman, M. (1987). Skin blood flow: A comparision of transcutaneous oximetry and lasers doppler flowmetry. European Journal of Vascular Surgery, 1, SISSIS.
- Baker, S.R., Stacey, M.C., Singh, G., Hoskin, S.E., & Thompson, P-J. (1992). Aetiology of chronic leg ulcers. European Journal of Vascular Surgery, 6, 245-251.
- Blair, S.D., Wright, D.D.I., Backhouse, CM, Riddle, E., & McCollum, CN. (1988). Sustained compression and healing of chronic venous ulcers. British Medical Journal, 291, 1159-1161.
- Bourne, I.H.J. (1986) Vertical drainage treatment of leg ulcers. In D. Negus & G. Jantet (Eds.), Phiebology '85. London: John Libbey & Co., Ltd.
- Brown, SJ. (1992), Tailoring nursing care to the individual client: Empirical challenge of a theoretical concept. Research in Nursing 6- Health, 15, 39-46.
- Bulstrode, C.J.K., Goode, A.W., & Scott, PJ (1987). Measurement and prediction of progress in delayed wound healing. The Journal of the Royal Society of Medicine, 80, 210212.
- Callam, M.J., Ruckley, C.V., Harper, D.R., & Dale, JJ. (1985). Chronic ulcération of the leg: Extent of the problem and provision of care. British Medical Journal, 290, 18551856.
- Cox, C.L. (1986). The interaction model of client health behavior: application to the study of community-based elders. Advances in Nursing Science, 9(1), 40-57.
- Ferraro, F.Z.F. (1980). Self-ratings of health among the oJd and the old-old. Journal of Health and Social Behavior, 21, 377-383.
- Gordis, L. (1979). Conceptual and méthodologie problems in measuring patient compliance. In R.B. Haynes, D.W. Taylor, & D.L.Sackett (Eds.). Compliance in health care. Baltimore: Johns Hopkins University Press.
- Harrison, J.A., Mullen, P.D., & Green, L.W. (1992). A meta-analysis of studies of the Health Belief Model with adults. Health Eduation Research, 7, 107-115.
- Janz, N.K., & Becker, M.H. (1984). The Health Belief Model: A decade later. Health Education Quarterly, 11,1-47.
- Johnson, M. (in press). Healing determinants in older people with leg ulcers. Research in Nursing and Health.
- Johnson, M. (1993). Physiological and psychosocial determinants of ieg ulcer healing in older Australians: Preliminary findings. Unpublished report to the Royal College of Nursing Australia.
- Johnson, M., & Miller, R. (in press). Measuring healing in leg ulcers: Appropriate instruments for practice and research. Applied Nursing Research.
- Johnson, W.C., & Patten, D.H. (1977). Predictability of healing of ischemie leg ulcers by radioisotopic and doppler ultrasonic examination. American Journal of Surgery, 133, 485-489.
- Lachman, M.E. (1986). Locus of control in aging research: A case for multidimensional and domain-specific assessment. Journal of Psychology and Aging, 1(1}, 34-40.
- Lau, R.R., & Ware, J.E (1981). Refinements in the measurement of health-specific locus-ofcontrol beliefs. Medical Care, U1 1147-1158.
- Marshall, GJ1J. (1991X A multidimensional analysis of internal health locus of control beliefs: Separating the wheat from the chaff? Journal of Personality and Social Psychology, 61, 483-491.
- Nagy, V.T., & Wolfe, G.R. (1984). Cognitive predictors of compliance in chronic disease patients. Medical Care, 22, 912-921.
- Nelzen, Q, Bergqvist, D., Lindhagen, A., & Hallbook, T. (1991). Chronic leg ulcers: An underestimated problem in primary health care among elderly patients. Journal of Epidemiology and Community Health, 45, 184-187.
- O'Leary, A. (1982). Self-efficacy and health. Behavior Research and Therapy, 23, 437-451.
- Pfeiffer, E. (1975). A short portable mental status questionnaire for the assessment of organic brain deficit in elderly patients. Journal of the American Geriatrics Society, 23, 433-441.
- Reid, D.W. (1984). Participatory control and the chronic-illness adjustment process. In H.M. Lefcourt (Ed.), Research with the locus of control construct (Vol. 3). New York: Academic Press.
- Rogers, R.W. (1975). A protection motivation therapy of fear appeals and attitude change. The journal of Psychology, 91, 93114.
- Rooke, T.W., & Osmundson, PJ. (1989). Variability and reproducibility of transcutaneous oxygen tension measurements in the assessment of peripheral vascular disease. AngMogy, 40, 695-700.
- Rosenstock, R.M., Strecher, V.J., & Becker, M.H. (1988). Social teaming theory and the Health Belief Model. Health Education Quarterly, 15, 175-183.
- Rotter, J.B. (1954). Social teaming and clinical psychology. New Jersey: Prentice Hall.
- Strecher, V.J., DeVillis, B.M., Becker, M.H., & Rosenstock, I.M. (1986). The role of selfefficacy in achieving health behavior change. Health Education Quarterly, 13(1), 73-91.
- Strickland, B-R. (1978). Internal-external expectancies and health-related behaviour. Journal of Consulting ana Clinical Psychology, 46, 1192-1211.
- Strickland, B.R. (1989). Internal-External control expectancies from contingency to creativity. American Psychologist, 44(1), 1-12.
- Wahl H.V. (1991). Dependence in the elderly from an interactional point of view: Verbal and observational data. Psychology and Aging, 6, 238-246.
- Wallston, K.A., & Wallston, B.S. (1981). Health locus of control scales. In H-M. Lefcourt (Ed.), Research with the locus of control construct: Assessment methods (Vol. 1) New York: Academic Press.
- Wallston, K.A., Wallston, B.S., & DeVUIiS, R. (1978). Development of the multidimensional health locus of control (MHLC) scales. Health Education Monographs, 6, 160170.
- Wallston, K.A., Wallston, B-S., Smith, S-, & Dobbins, CJ. (1987). Perceived control and health. Current Psychological Research ff Reviews, 6(1), 5-25.
- Weisz, J.R., Rothbaum, P.M., & Blackburn, T.C(1984). Standing out and standing in. The psychology of control in America and Japan. American Psychologist, 39, 955-969.
- Winter, G.D. (1971). Healing of skin wounds and the influence of dressings on the repair process. In KJ. Harkiss (Ed.), Surgical dressing and wound healing symposium. Bradford, UP: Crosby Lockwood.
- Wurtele, S.K., & Maddux, J.E. (1987). Relative contributions of protection motivation theory components in predicting exercise intentions and behavior. Health Psychology, 6, 453-466.
Docf or-Dicuynosecf Health Problems In Sample
Distribution of Ulcer Sizes in the Sample
Group Comparisons of Healing Rafe and Response Efficacy Beliefs