Elders and their families are faced with finding alternative long-term care solutions when home care is no longer adequate. There are often economic constraints on the family, limiting the choice of long-term care residential homes. Unfortunately, the prevailing assumption is that affordable institutions do not provide care with a focus on maintaining or enhancing the elder's quality of life.
This study demonstrates that nursing care that enhances quality of life can be provided to all nursing home residents without additional cost to either the institution or resident. The nursing staff focuses on the holistic needs of the elderly residents instead of orienting their care around procedures and tasks. This approach gives the patients greater control and more active involvement in their own care. The ultimate result is greater patient wellbeing. This approach to nursing care raises staff and management issues that are critical to support the institutionbound elder's dignity and life quality.
The literature confirms the prevailing assumption that residents of nursing homes often have limited or nonexistent control over their everyday activities. The more limited the institutional resources (ie, staff, activities), the more restricted the patients' options. The restriction of personal choice includes decisions about eating, rising and retiring for the day, dressing, bathing, having a cocktail, and, most certainly, sexual behavior. Indeed, the resident may not have had a choice of whether to go to a nursing home.
Schultz and Brenner reviewed the literature concerning relocation of the aged as a stress factor, and proposed a model to explain individual response (mortality) to this stress.1 They argued that the result is largely determined by perceived predictability and controllability of the events surrounding a move; and differences in controllability between pre- and post-relocation environments. Specifically, involuntary relocation is associated with high rates of deterioration and death.
Schultz also determined that predictable and controllable positive events had a powerful positive impact on the institutionalized elder's well-being.2 Visits from undergraduate students (positive event) were offered to elders in two forms: the time and length of visit were controlled by the elder; and the time and length of the visit was predictably set by the researchers. A follow-up study revealed that the groups that benefited most from the treatment, suffered most from the loss of student visits.3 The groups that had no visits or had random visits remained stable over time.
Langer and Rodin conducted a field experiment to assess the effects of enhanced personal responsibility and choice on a group of nursing home residents.4 The experimental group was encouraged to assume more responsibility for their personal lives and they were given a house plant with care instructions. The control group, selected from another floor, were not encouraged to increase responsibility for their Uves, and although they were provided with plants, the staff assumed care of the plants. Health and wellbeing ratings and behavioral measures showed significant improvement in the experimental group, with maintenance of these positive outcomes in a followup study.5 A similar study using two different nursing home settings was reported by Mercer and Kane, and the outcomes were positive for the experimental group.6
Chang looked at the interactive effects of generalized expectancy of control and perceived situational control on the morale of institutionalized elders.7,8 She found that subjects who perceived daily activities to be selfdetermined scored higher on morale, regardless of the locus of control category. Pohl and Fuller also found that "choice within an institution" was a major contributor to morale in a sample of 50 residents in a home for the aged.9 Ryden similarly found that perceived control was a key variable in 113 residents' morale scores, with the association being stronger among the skilled care subjects than those in intermediate care.10
Moos further studied choice and control in community care settings for older people.11 He found that residents who were functionally able and were women were more likely to display independence and greater social functioning when in facilities higher in choice and control, Berkowitz, Waxman, and Yaffe compared residents in a self-help senior residence with those in more conventional elder residences.12 As predicted, self-help residents scored significantly higher on control, self-esteem, and social involvement.
Avorn and Langer found that with the institutionalized elderly, assistance from the staff reduced the subjects' ability to perform a simple psychomotor task themselves.13 They concluded that excessive infantilization of residents and overly intrusive help in selfcare beyond clinical requirements can lead to "learned helplessness," with further disability.
Elderly adults' desire and expectancy for control was the focus of two studies by Smith et al.14 They found that age was related to generalized desire for control and to control over health matters. For health-specific measures at older ages, belief in internal control and desire for control are lower, and belief in control by powerful others is higher. The study did not find differences in generalized desire for control across ages. Wishing for control over daily activities is not necessarily related to control over health care. Alternatively, the age-associated finding may not be due to actual age, but may be a cohort effect.
Cox, Miller, and Mull sought to identify indicators of self-determination and competency (intrinsic motivation) in the health behavior of elders.15,16 Being female, having a more positive sense of well-being, higher education, and a better perception of one's health status contributed significantly to feelings of competency and self-determination in elders relative to their health behaviors. Being male or white was associated with decreased feelings of competency and selfdetermination. Elders who perceived themselves as healthier than their peers had higher levels of health-related intrinsic motivation.
Despite multiple descriptions of the impact of elders' choice and control on their health and well-being, little research has described long-term care interventions that reorient and stabilize staff around residents' needs for choice and control. To this end, the Quality of Life Nursing Care (QLNC) model was developed by the second authoi; implemented, and evaluated.
DESCRIPTION OF THE QLNC MODEL
The QLNC model encompasses four separate but inter-related components: nursing staffs focus on the resident's choice and control to the greatest extent possible; permanent assignment of nursing assistants to residents; provision of case-managed nursing care; and permanent resident-centered scheduling.
The licensed staff and nursing assistants caring for residents in the QLNC model do not focus exclusively on "body" care, but instead expand their focus to include important quality of life care considerations. The goals of nursing action are optimal resident functioning, social and psychological wellbeing, health maintenance and promotion, and control over decisions. Inservice education programs for nursing staff specifically target the provision of quality of life nursing care. The facility management supports staff behavior based on these goals through appropriate policies and staff evaluation.
Each resident is permanently assigned full-time nursing assistants to provide care on the day and evening shifts. One nursing assistant provides care 5 days per week, and another 5 evenings per week. When these nursing assistants have days off, a relief nursing assistant assumes responsibility for that specific group of residents on that day and that shift. The full-time nursing assistants work the same days and hours as the case manager nurse; and the relief nursing assistants work the same hours as the relief nurse. The same licensed nurse permanently supervises the same nursing assistants. The staff informs residents and their families about which nursing assistants are caring for them on each shift, and whenever there is a change.
Through permanent assignments, each resident has one case manager (CM) nurse who is responsible for the management and integration of nursing care. The CM nurse has the authority to initiate interdisciplinary assessment and care planning conferences and supervises the related documentation. In addition, the CM nurse monitors and evaluates implementation of the care plan and the residents' responses to the care. Finally, the CM nurse provides or supervises care for all of the unit's residents. Each unit (group of residents) has a team of one CM nurse and one or more nursing assistants working days and another team working evenings during a 24-hour period. One or the other of the CM nurses for the unit is always on duty. When one of the CM nurses has a day off, a relief nurse assumes the routine responsibilities for that day and shift. Necessary case management responsibilities are assumed not by the relief nurse, but by the unit CM nurse on the other shift. Residents, their families, their physicians, facility management, and others who work with the residents are explicitly aware of who is serving as the CM nurse.
All staff are assigned to permanent days and hours equally distributed across all days and across the day and evening shifts. Three times per week, full-time day and evening staff have an hour of overlap time. During this hour, the two staff plan, coordinate, and evaluate care with and for specific residents on the unit. They also use these hours for inservice education, unit management, and team building across shifts. Fifteen minutes of overlap time is scheduled between all other day and evening shifts. All staff use the time to coordinate daily activities on behalf of the residents. Starting time for the day shift (ie, 6 AM versus 7 AM) is based on the desired rising time for most residents on the unit; similarly, the finishing time for the evening shift (ie, 10 PM versus 11 PM) is based on the desired bedtime for most residents. Each day, care is taken to have on duty either the unit's full-time day staff or the unit's full-time evening staff.
A longitudinal nonequivalent control group design was used to test the effects of the experimental QLNC model on quality of life outcomes of elder residents in long-term care facilities; facility management's attitudes and commitment toward quality of life nursing care; and full-time licensed and unlicensed nursing staffs attitudes toward and perception of administrative commitment relative to quality of life nursing care. Two intermediate care units were selected from a nursing home; one was randomly designated as the experimental unit and the other as the control unit.
A 400 to 500 bed proprietary nursing home, located in a blue-collar community, provided two intermediate patient care units (one experimental, one control) for this study. Each unit had 78 beds with 90% occupancy on the experimental unit and 83% occupancy on the control unit pre-intervention. Approximately 90% of the residents were Medicaid patients. At the end of the study, there were 97% and 95% occupancy rates, respectively, on the experimental and control units.
The QLNC Model Intervention
The QLNC model was implemented for 6 months on the experimental unit. Specifically, nursing assistants were permanently assigned to residents; case manager nurses for each resident were identified; shift schedules were altered to accommodate the unit residents' preferred wake/sleep patterns (eg, 6 AM to 2 PM; 2 PM to 10 PM); and licensed nursing staff received 10 weekly 1 -hour sessions containing content that emphasized resident control and choice. Unlicensed staff did not receive the 10 weekly inservice sessions. This was done deliberately because the investigators wanted to approximate as closely as possible the realities of the institution. Right or wrong, time and staff constraints dictated that inservice opportunities were primarily reserved for licensed staff. In addition, the daily team meetings between licensed and unlicensed staff were intended to be a vehicle for conveying some of the licensed staffs learning to the unlicensed staff In collaboration with the investigators, master's and doctorally prepared gerontological nursing faculty developed and taught the inservice sessions.
Subjects and Sampling Procedures
The target populations for this study included facility residents who were cognitively intact, 60 years of age or older, and had lived in the facility at least 6 months; full-time licensed and unlicensed nursing staff who were providing direct care on days or evenings to the patient sample selected; and facility management staff, including the director and supervisory nursing personnel. Sample size for the QLNC evaluation was based on an estimated 10% loss of subjects, a 10% anticipated refusal rate (by resident or resident's family or significant other), as well as on the analytic procedures that are sensitive to sample size.
A systematic random sample of 23 experimental and 22 control residents was selected from the patient census lists provided by the unit. The final postintervention sample (residents completing both the pre- and post intervention measures) consisted of 39 residents (21 experimental; 18 control).
All administrators (N = 5) were included in the study. All day and night licensed nursing personnel (3 experimental; 2 control) and all day and evening nursing assistants from both the experimental (n = 6) and control units (n = 4) were included in the study. Because the licensed nursing staff felt that the night shift staff was subject to change and had minimal contact with residents, night shift nursing assistants were not included.
After the random sample of residents had been selected, interviewers approached the elders individually in their rooms to explain the study purposes, obtain the resident's consent to participate, and arrange a time to complete the survey instrument. If a resident did not wish to be in the study, the interviewers made no further attempt to pursue cooperation. Nursing students (master's and doctoral level) who routinely participated in similar projects conducted the study interviews. Interviewers' inter-rater reliability was established by the principal investigator at a 0.80 correlation standard. All unit staff, administrators, and nursing service administrative personnel responded to a self-administered questionnaire appropriate to their roles.
Measures for Residents
Decisional control was measured by the Locus and Range of Activities Checklist, which consists of 30 items representing elements of daily life (eg, who to sit with at meals, what time to eat meals, what time to go to bed).17 For each item, the resident indicated on a three-point scale the amount of choice available regarding that item (Cronbach's alpha pre-intervention = .87; postintervention = .92).
Physical functioning was assessed by two scales. The Self-Management Scale designed by the senior author for a previous study presented residents with 10 items on which they indicated to what extent they could perform that function (eg, walking, bathing, dressing) in a reasonable amount of time without assistance.18 The Cantrill-like ladder format provided a visual analog for the resident (Cronbach's alpha pre-intervention = .91; postintervention = .89). The number and kind of symptoms that might be functionally limiting were measured by the Symptoms Index19 (Cronbach's alpha pre-intervention = .83; postintervention = .81).
Psychological well-being was measured by the General Well-Being Schedule.20,21 Eighteen items divided into six subscales measured health worry, energy level, satisfying-interesting life, depressed-cheerful mood, emotionalbehavioral control, and relaxed versus tense-anxious affect (Cronbach's alpha pre-intervention = .87; postintervention = .88).
Size of the social network was measured by Bennett's Past Month Index.22 This index considers the number of role relationships in which the respondent is actively involved.
Loneliness was measured by two single item responses on a five-point Likert scale. One item appeared early in the interview, and the second appeared toward the end. The investigators decided on the single item measure in lieu of the loneliness scales because the scales are often confounded with measures of anomie, alienation, alienation via rejection, authoritarianism, and depression.23,24
Social activity items for this study examined the extent to which the resident was socially engaged, both inside and outside the facility. Activities included church attendance, shopping, clubs, recreation, other social dining, interactive games, television, needlework, reading, political activity, and the last time the respondent was outside of the facility.
A self-perceived health status assessment was derived from the crosstabulation of two items to create an index: Would you say your health is excellent, good, fau; or poor? In comparison to others your age, is your health better, the same, or worse? According to Kviz and Flaskerud, this method produces an elegant measure of self-health assessment.25 Self-health assessment has been repeatedly demonstrated to correlate significantly with the medical assessment.26,27
Satisfaction with care was measured by 22 items that address location, amenities, services, routines, and privileges within the nursing home. Respondents indicate that (O) they are not at all satisfied, (1) somewhat satisfied, or (2) very satisfied (Cronbach's alpha preintervention = .78; postintervention = .70).28
Measures for Staff and Administrators
Staff and administrators' attitudes were examined using the Quality of Life Questionnaire developed by the co- investigator.29
Nursing staff were asked about their attitudes toward resident or family control of choice in daily activities (12 items). Five-point Likert response options ranged from facility staff should make most decisions (1) to residents or family should make most decisions (5) (Cronbach's alpha pre-intervention = .68; postintervention = .65).
Nursing staff were asked about their attitudes toward holistic nursing care as they related to patients, including food and eating; resting, sleeping, and awakening; toileting; bathing and grooming; sensory experiences; physical activity; cultural activity; companions; environment; therapies; medications; and treatment (Cronbach's alpha pre-intervention = .73; postintervention = .77).
Staffs perception of administrative commitment to holistic nursing care was evaluated by 10 items (Cronbach's alpha pre-intervention = .71; postintervention = .81).
Nursing staff were asked about their attitudes toward quality of life versus prolongation of life. Likert response options included a range from 1 to 5 (Cronbach's alpha pre-intervention = .63; postintervention = .50).
Staff perceptions about administrative commitment to support quality versus prolongation of life were tapped through five items (Cronbach's alpha pre-intervention = .63; postintervention = .50).
Staff were polled on four items designed to measure their attitude toward resident-centered staffing. Items included their responses to case management and permanent assignment of staff to residents (Cronbach's alpha preintervention = .84; postintervention - .87).
Staffs perception of administrative commitment to resident-centered staffing was elicited through 13 items, tapping specifics about case management, shift schedules, and day/night schedules (Cronbach's alpha pre-intervention = .78; postintervention = .70).
Perceived quality of care was measured by 1 1 items addressing interpersonal, structural, and environmental aspects of quality.
Administrative staff responded to the same items as did other staff, but they were phrased from the administrator's viewpoint (Cronbach's alpha pre-intervention = .76; postintervention = .69).
Reliability of the Measures
Cronbach's alphas less than .65 on either the pre- or postintervention measures defined the instrument as unsuitable for use in subsequent analyses. This level, in contrast to Nunnally's usual standard of .70 for newer scales,30 was selected due to the potentially joint effects of both sample size and small number of items per scale. Applying this criterion, only two scales were eliminated - Staff and Administrator Attitude toward Quality of Life and Perceived Administrative Commitment to Quality of Life.
The resident sample in both units was predominantly white, widowed, and from blue collar occupations. Residents on the experimental unit were younger, better educated, had higher incomes, entered the nursing home at an earlier age, and had resided at this nursing home longer than the control residents. No significant differences among staff existed between units.
Intervention Effects for Residents
Residents on the experimental unit reported significant increases in control and choice and well-being from pre- to post-test, whereas residents on the control unit did not. Residents on both units reported a significant decline in self-care competency, and residents on the experimental unit reported a decline in health status.
Because age, education, income, age when entered nursing home, and length of nursing home residence varied significantly before the intervention between the experimental and comparison groups, these variables were statistically controlled when examining the postintervention resident measures. The demographic variable difference had no impact on the postintervention measures. True unit differences postintervention existed in health status; the experimental unit residents reported a lower health status than those in the control group (ANCOVA, F (1/ 37)=4.08,P≥.05).
For residents on the experimenta] unit, general health status declined (f (19)= -2.98, P^.OOSX choice/control increased (t (20) = 3.84, P^.OOIX selfcare competency decreased (t (19)= -2.12, P=≤.05), and general well-being increased (t (17) = 3. 25, P=0.005). One significant change was observed among the comparison unit residents; they, too, reported decreased self-care competency (t (16)= -2.39, P≤ .03).
Intervention Effects for Staff
Experimental and control staff did not differ significantly on any of the attitudinal measures pre-intervention. Postintervention, the experimental unit staff expressed a more positive attitude toward resident control and choice (ANCOVA, F (l/9) = 5.29, P*£.03) and perceived their quality of care to be higher (ANCOVA, F (l/9) = 6.82, P^.Ol) than did staff on the control unit. Because the unlicensed staff sample was nearly twice that of the licensed sample, the groups were separated and then compared. Only one major difference between the licensed and unlicensed staff emerged postintervention: licensed staff were more positive than unlicensed staff postintervention in their attitudes toward resident control/choice (t (9) = 2.3, P*s.02). When only the experimental licensed and unlicensed staff were compared, licensed staff again were more positive toward resident choice (t (5) = 2.1, P*s .04).
Intervention Effects for Administrators
Administrators had only one significant change from pre- to postintervention: they expressed even greater commitment to resident-centered staffing (t (2) = 6.30, P=S .008). Positive, but nonsignificant, postintervention trends were noted in the administrators' commitment to holistic nursing care and perceived quality of nursing care.
Despite onl a 6-month triai of the QLNC model, a number of significant findings can be directly attributed to the new nursing care approach. Resident choice and control and general well-being were increased on the experimental unit, even in the face of a declining health status and self-care competency. Experimental unit staff were more positive in their attitudes toward resident choice and control postintervention than were staff on the control unit. The effect of the QLNC model on administrators' attitudes was positive; they expressed greater commitment to resident-centered staffing postintervention.
That the QLNC model is a viable option for nursing homes is tentatively established. Despite the fact that staff were limited and the nursing home served primarily Medicaid recipients, positive changes in the patients occurred on the experimental unit. The impact of the QLNC model might have been greater had the intervention period been extended for staff preparation and orientation to the model on the experimental unit. Allotting time to deal effectively with the process of change, particularly for unlicensed staff, would potentially increase the potency of the staff re-education.
The licensed staff were the primary recipients of the inservice education sessions that focused on QLNC model components. They were also more positive toward resident control/choice than were the unlicensed staff. Although licensed staff were involved in direct client care, their resident contact was much more limited than that of the nursing assistants. Nursing assistants, no doubt, received indirect benefits of the educational offerings, but their numbers alone may have diminished the effects of the intervention. To increase the potential effectiveness of the QLNC model, subsequent efforts will most certainly design specific educational offerings for both the nursing assistants and the licensed personnel.
As with any field study, a number of factors beyond investigator control could have influenced the results of the study. Mortality and history were certainly significant threats to internal validity; six residents died or were transferred to other units (one resident was transferred from the experimental unit to the control unit). External validity is limited in that only one nursing home was examined. Subsequent studies will focus on multiple nursing homes and include a closer examination of structural and organizational variables that might advance or impede the success of the QLNC model.
Despite a better demographic profile of the experimental group, the decline in health status and self-care competency was similar in both groups. This suggests that physiological factors and time - not the intervention - were probably responsible for a decrease in health status and associated self-care ability. More importantly, the fact that well-being on the experimental unit increased despite declining health status strengthens the case that the QLNC model intervention was effective at no extra cost to personnel.
There is beginning evidence to support the use of the QLNC model as an effective approach to optimize resident choice, control, and well-being. Data are offered to identify important attitudes and perceptions of staff and administrators that potentially may affect both the implementation and ultimate effectiveness of such an intervention approach. Subsequent testing with larger sample sizes, variable settings, and longer intervention trials should definitively address the value of the QLNC model for nursing home care regardless of patient and faculty resources.
- 1. Schultz R, Brenner G. Relocation of the aged: A review and theoretical analysis. J Gerontol. 1977; 32:323-333.
- 2. Schultz R. Effects of control and predictability on the physical and psychological wellbeing of the institutionalized aged. J Pers Soc Psychol. 1976; 33:563-573.
- 3. Schultz R, Hanusa B. Long-term effects of control and predictability-enhancing interventions: Findings and ethical issues. Journal of Personality and Psychology. 1978; 36:li94-1201.
- 4. Langer EJ, Rodin J. The effects of choice and enhanced personal responsibility for the aged: Field experiment in an institutionai setting. J Pers Soc PsychoL 1976; 34:191198.
- 5. Rodin J, Langer E. Long-term effects of control-relevant intervention with the institutionalized elderly. J Pe rs Soc Psychol. 1977; 35:897-903.
- 6. Mercer S, Kane RA. Helplessness and hopelessness among the institutionalized aged: An experiment. Health Soc Work. 1979; 4(0:91-116.
- 7. Chang BL. Perceived situationai control of daily activities: A new tool. Res Nurs Health. 1978; 1(4):181-188.
- 8. Chang BL. Generalized expectancy, situationai perception, and morale among institutionalized aged. Nurs Res. 1978; 27:316324.
- 9. Pohl JM, Fuller SS. Perceived choice, social interaction, and dimensions of morale of residents in a home for the aged. Nursing in Research and Health. 1980; 3:147-157.
- 10. Ryden MB. Morale and perceived control in institutionalized elderly. Nurs Res. 1984; 33:130-136.
- 11. Moos RH. Environment choice and control in community care settings for older people. Journal of Applied Social Psychology. 1 98 1 ; ll(l):23-43.
- 12. Berkowilz MW, Waxman R, Yaffe L. The effects of a resident self-help model on control, social involvement, and self-esteem among the elderly. Gerontologist. 1988; 28:620-624.
- 13. Avom J, Langer E. Induced disability in nursing home patients: A controlled trial. J AmGeriatrSoc. 1982; 30:397-400.
- 14. Smith RA, Woodward NJ, Wallston BS, Wallston KA, Rye P, Zylstra M. Health care implications of desire and expectancy for control in elderly adults. J Gewntol 1988; 43(1):P1-P7.
- 15. Cox C, Miller E, Mull C. Motivation in health behavior: Measurement, antecedents, and correlates. ANS. 1987; 9:1-15.
- 16. Cox C. Health self-determinism index. Nurs Res. 1985:34:177-183.
- 17. Hulicka I, Morganti J, Cataldo J. Perceived latitude of choice of institutionalized and noninstitutionalized elder women. Exp Aging Res. 1975; 1:27-39.
- 18. Cox C. TTie interaction model of client health behavior Application to the study of community-based elders. ANS. 1986; 9:40-57.
- 19. Sullivan JA, Armignacco F. Effectiveness of a comprehensive health program for the well elderly by community health nurses- Nurs Res. 1979; 28:70-75.
- 20. Fazio A. A Concurrent Validation Study of the NCHS General Well-Being Schedule. Washington, DC: Department of Health and Human Services; 1979.
- 21. HimmeUarb S. Age and sex differences in mental health of older persons. J Consult ClinPsychol. 1984; 54:844-856.
- 22. Bennet R. Aging, Isolation and Mental Illness in Old Age. New York: Van Nostrand Reinhold Co; 1980.
- 23. Francis G. Loneliness: Measuring the abstract. International Journal of Nursing. 1976; 13:153-160.
- 24. George L, Blearon L. Quality of Life in Older Persons: Meaning and Measurement. New York: Human Sciences Press; 1980.
- 25. Kviz FJ, Flaskerud JH. An evaluation of the index of medical under service: Results from a rural consumer survey. Med Care. 1984; 22:877-889.
- 26. Ferrare K. Self-rating of health among the old and the old-old. J Health Soc Behav. 1980; 21:377-383.
- 27. Cockerham W, Sharp K, Wilcox J. Aging and perceived health status. J Gerontol. 1983; 38:349-355.
- 28. McCafftee K, Harkins E. Final Report for Evaluation of the Outcomes of Nursing Home Care. Seattle, Wash: Battette Human Affairs Research Centers; 1976.
- 29. Kaeser L. Quality of Life Nursing Care: An Experimental Trial. Chicago: University of Illinois at Chicago; 1986. Unpublished manuscript.
- 30. Nunnally JC. Psychometric Theory, 2nd ed. New York: McGraw-Hill Book Co; 1978.