Advances in scientific knowledge and technology have added years to the human life span and have contributed to the rapidly increasing elderly population. These phenomena have escalated the incidence and degree of chronic illness and intensified the necessity of examining geriatric health care requirements (Robb, 1984). Limited health care resources exacerbate this quandary by creating a challenge for creators of health care policy. To counter this difficulty, researchers have spotlighted gerontological research as a nucleus for investigation-with a goal of allocating the scarce health care resources (Williams, 1984).
Nursing research guides nursing practice. The existing body of knowledge to guide nursing practice for the elderly patient is limited. Thus, an issue for nursing has been created in the face of a paucity of health resources and heightened health care requirements. Nurse scientists, educators, and clinicians have begun gerontological investigations to address this issue (O'Leary, 1990). The purpose of this article is to identify issues, concerns, and problems specific to research of the aged through a review of the literature.
STUDY NECESSITIES/RESEARCH QUESTIONS
Research questions should be important to the researcher and gerontological nursing. Available funding, possibilities for replication, and ethical issues must be considered.
When identifying a research question for study, a scientist must be politically wise and consider areas that are being funded at local, state, and federal levels (Brimmer, 1987; Hinshaw, 1988). Gerontological areas requiring development with priority for federal funding include longterm care for the frail elderly; iatrogenic complications; quality and continuity of care; self-care issues; and coping and adaptation of patients/ caregivers (Hinshaw, 1988). General questions to consider when contemplating an area of study include the following:
* Should all aspects of geriatricpatient needs be investigated?; and
* Should efforts be concentrated on two or three specific areas with consistent replication, so that findings can begin to make a difference?
Replicated gerontological studies are limited (Brimmer, 1987; Engle, 1990), although the arena of gerontological research is broad. Thus, original research is disseminated over a considerable area without substantial support from replicated findings for many constructs of interest. Perhaps researchers, including master's and doctoral students, should consider replicating research for their investigations. Considering the number of theses and dissertations, replication would have the potential to greatly advance nursing science.
Ethical Aspects of Research Questions
Historically, mortality has been used as an outcome or as the dependent variable in geriatric clinical research (Williams, 1984). In formulating a research question, scientists must consider whether death (or illness) is a negative outcome. Ethically, the quality of Ufe issue arises in opposition to the quantity of life. Which one outweighs the other - a better life or a longer life? On which one should scientific investigations focus?
One example of a research question with an ethical component involves decision making - a choice between freedom of the individual versus what others (significant others and health care professionals) think is best for the patient. The following is a situational example:
In preparing for discharge from the hospital, a mentally competent elderly patient requests that she be allowed to return home to die. She is chronically ill and her mobility is restricted. She realizes that monies are not available for round-theclock home health care and that family members are not able to, or willing to, care for her on a continual basis.
Hospital discharge planners, nurses, and physicians are reluctant to allow the elderly individual this freedom of choice; they decide, with the consent of family members, that the patient will be placed in a nursing home. Although cognizant of the situation, the patient is not completely aware of her decision-making rights. Nor does she have the energy to contest the decision. Therefore, she unhappily succumbs to the desires of her family and the "trusted" health care professionals. She will spend her remaining days in a nursing home.
In addition to ethical implications regarding decision making, legal and financial ramifications would most likely affect the decision for nursing home placement. Health professionals may have chosen to institutionalize the patient for her safety and protection. At the same time, however, professional liabilities were minimized from possible patient injuries in a less safe environment, such as the patient's home. The family may feel relieved from guilt and responsibility as the nursing home staff assumes the care of their loved one. The family may also experience economic relief if Medicaid covers the cost of the nursing home, realizing that continuous home health care is not a reimbursed expense.
Unfortunately, this situation is common among the elderly, especially as hospitalization and illness exacerbate vulnerability (Ryden, 1984). This situation raises the issue of desired outcome. Was the outcome favorable in that the patient was released from the hospital? Was the outcome unsatisfactory because the patient needed nursing home placement? Was the outcome immoral because the decisionmaking rights of the elderly individual were not considered? To what extent did our health care reimbursement system affect the family decision for nursing home placement? Were needs met? Whose needs were met?
Many obstacles in gerontological research represent methodological problems - including sampling, instruments, and design. Informed consent also poses a dilemma, sometimes presenting an impasse before the study begins.
In general, large samples are necessary to counter the effect of having an increased variance from heterogeneity (Burns, 1987; Polit, 1987). The elderly are not a homogeneous group of people (Brooke, 1988; Gueldner, 1989; Williams, 1984). The variance of biological, functional, and psychosocial characteristics increases with age, which makes measurement less precise.
What constitutes normality of the aged? Studies of the elderly frequently include people over the age of 65, as if homogeneity of people 65 and older could be assumed (Williams, 1984). In attempts to homogenize the elderly, older individuals have been classified into biologic, social, and psychologic ages (Birren, 1985); and into young-old (60 to 74 years), middle-old (75 to 84 years), and old-old (85 years and over) classifications (Robb, 1984).
Establishing a sampling frame for randomization with elderly populations is often problematic. Centralized listings of older or retired persons often are not available. Social Security offices have a listing; however, the names are confidential and cannot be released. Additionally, not everyone over the age of 65 receives these benefits (Gueldner, 1989). Limited access to subjects and nonrandomization leads to unrepresentative sampling and systematic errors (Burns, 1987). One positive aspect of sampling with certain elderly groups, such as residents of nursing homes, is that these people represent a stationary sample, easing subject access.
Consent forms must be signed by each subject prior to initiating observation or treatment (Burns, 1987). Obtaining these signatures is often a barrier in research with the aged (Duffy, 1989). Consent forms should be brief, clear, and nonthreatening; however, any risks must be explicated (Burns, 1987). Unfortunately, acknowledging risks can generate fear in nursing home subjects. Gueldner (1986) reported that some elderly patients who had previously agreed to participate in a study declined when confronted with a consent form to sign. Older adults, particularly in nursing homes, may refuse to sign consent forms because family members have warned them not to sign any type of form (Duffy, 1989).
It is generally accepted that intellectual and cognitive abilities decrease at varying rates with age. Gerontological researchers using interviews for data collection must consider the elderly subject's gross mental competency and intellectual functioning. Mental status exams should be used prior to patient involvement to determine eligibility to participate in the study. Using such a measure would preclude inaccurate responses that are due to decreased mental acuity; such responses could invalidate study results. Using succinct mental status exams will reduce interview length and fatigue of the elderly interviewee (Pfeiffer, 1975).
Self-administered questionnaires are popular for data gathering because of economical advantages and because they make having a large sample a realistic possibility. However, using questionnaires in elderly populations has significant disadvantages that cannot be overlooked. Visual, mental, cognitive, and fingerdexterity deficits may affect the accuracy with which questions are answered. The use of large, bold, black type on bright yellow or orange paper to accentuate the contrast - and additional space between items - may lessen the visual burden and increase accuracy. Graduated pages or finger tabs may ease problems with decreased finger dexterity and help avoid skipped pages (Gueldner, 1989).
Researchers often have difficulty distinguishing between data that is necessary and data that is interesting when devising questionnaires. The inclusion of unnecessary data tends to lengthen the form and results in not obtaining the most important information (Kirchhoff, 1991). Making the distinction between what the researcher requires and what the researcher wants is crucial to simplifying and shortening the process.
Valid and reliable instruments to measure outcomes of nursing interventions are lacking in elderly populations (Engle, 1990). Instruments must be validated for a specific population (Burns, 1987). An instrument validated on the young may not be valid for the elderly (Williams, 1984); this can be perplexing, because validity and reliability are typically established on younger populations (Gueldner, 1989).
Additional contributors to inaccurate measurement in the elderly include vision /hearing deficits and test anxiety. The lack of formal education in the elderly has been acknowledged, but was found to be higher in institutionalized patients (Gueldner, 1986). This is important for longterm care researchers to consider; many instruments have been created at a high-school reading comprehension level (Gueldner, 1989).
In one investigation, elderly subjects were more receptive to interviews than questionnaires, behavioral responses than verbal responses, and dichotomous answers than rating-scale options. Personal interaction and ease with which the elderly subjects responded were perceived as important (Brooke, 1988), and must be taken seriously when considering a design for gerontological research.
A limited number of research designs have been used to study elderly subjects. Although the limited number does not necessarily decrease sophistication of the research, gerontological nursing will not fully develop until all research strategies are used (Whall, 1990). Research studies are analogous to educational strategies. Nurse educators use a variety of teaching strategies to more fully capture the interest of all student learners (Partridge, 1983). Likewise, nurse scientists must use diverse research designs to holistically apprehend health care needs of the elderly and to generate and test gerontological nursing theory.
Nurse gerontologists cannot continue the research debate over quantitative /qualitative data because both types of data are needed to advance science. Carper (1978), and Chirm and Kramer (1991) enlighten and remind nurses that we do not know (to have or gain nursing knowledge) solely by one method. Empirical, personal, ethical, and aesthetic components all contribute to nursing knowledge. Quantitative data may most easily test propositions and hypotheses in an empirical quest. However, qualitative methodology may be more effective in generating theory and capturing personal, ethical, and aesthetic components of nursing knowledge. Both types of data are necessary for comprehensive research of the aged.
Descriptive designs have been most frequently used in gerontological studies (Engle, 1990; O'Leary, 1990). One strategy for moving from descriptive approaches to more sophisticated designs is the use of metaanalysis (Engle, 1990). Meta-analysis can strengthen theory testing in gerontology by pulling together relationships among constructs from many studies - and then following-up with testing of those linkages. This would permit a movement from descriptive designs to quasi- or experimental strategies. A prerequisite to meta-analysis is having four or five articles on a particular topic. Considering the recent surge in gerontological research, this would open up several subtopics for quantitative analysis.
Analogous to meta-analysis is the synthesis of findings from selected areas that have a modest but significant body of knowledge (Wells, 1988). Examples of such areas include falls, pressure sores, and resocialization interventions. Instead of recreating the wheel, models to direct nursing practice could be derived from syntheses of existing knowledge. Theoretical models not typically used in gerontological nursing research (Engle, 1990) would offer structure and guidance to the complex phenomenon of aging (Brimmer, 1987).
The ethnographic or ethnonursing design is becoming an accepted strategy for identifying health care needs of the elderly. This approach is especially useful when little is known about the subject and when the goal is to gain meaning within a cultural context of elderly individuals (or to make cultural comparisons) (Cameron, 1990). For example, cultural immersion in a long-term care facility could identify the meaning of institutionalization for elderly residents. Ethnography may offer insight into gerontological study. Moreover, it would bypass methodological problems often encountered in a quantitative approach.
Most studies across gerontology disciplines are cross-sectional (O'Leary, 1990), which is most likely due to time savings and economical advantages. A naturally occurring problem with these designs is determining whether age or cohort is responsible for cause-and-effect relationships. These confounds must be separated and examined to avoid erroneous results. This problem can be largely avoided with the use of longitudinal studies because of the ability to examine changes in phenomena over time (Kovach, 1989).
Historically, longitudinal observations in geriatric populations have been considered important because of the necessity to follow patients with chronic diseases over time. Unfortunately, this necessity has been countered by the disadvantages of longitudinal research, which include major expense and time investments. Senior researchers may not have time to invest in longitudinal studies. Younger researchers often choose not to invest their time because of political ramifications, such as the need to publish frequently and quickly for promotion, tenure, and academic survival.
STAFF INVOLVEMENT AND EDUCATION
Staff involvement is of particular importance to the research process in long-term care facility studies (Brooke, 1988). Unlike acute care settings, staffs in long-term care facilities become attached to and even possessive of their patients. Therefore, the researcher must establish a trusting relationship and rapport with the staff long before patient contact. Maintaining harmony with the staff is an additional challenge because of high nursing turnover rates in long-term care. Consequently, providing information about the study must be a continual process.
NURSING IMPLICATIONS FOR RESEARCH AND PRACTICE
O'Leary (1990) reviewed gerontological journals over a 2-year period to assess the contribution of nursing and allied health research to nursing practice with elderly patients. Although the quantity of nursing research was limited, findings indicated that nearly 80% of the studies had the potential for useful contributions to nursing care. It was noted that the nonnursing studies were also applicable to nursing care as defined by the American Nurses Association parameters for gerontological nursing and could plausibly stimulate additional nursing research.
Nursing literature definitively states disapproval regarding nurses using borrowed knowledge. Perhaps it is time to rethink this common stance. Nursing gerontological research is currently limited. Furthermore, nurses do not and cannot practice in isolation; we serve patients in collaboration with a multidisciplinary health care team. Professionals must unite, using research available from all disciplines to strengthen gerontological science.
One idea or study, whether from nursing or allied health professionals, has the potential of leading to another. A research study in one nursing home had a high level of nursing staff involvement; being actively involved in this process stimulated inquiry by the nursing staff into other daily nursing practice problems (Brooke, 1988). Helping each other to help the elderly patient is what gerontological research is all about.
The deficiency of knowledge regarding the health care needs of the elderly has been acknowledged, and the scientific quest has begun. Multiple challenges are encountered with investigation of the elderly, who represent a diverse, heterogeneous population. Innovative efforts are necessary to obtain informed consent from subjects and to elicit cooperation from staff members to access and maintain subjects. Mental status exams should be used with elderly subjects if interviewing techniques are used. All instruments used should be reliable and validated specifically for elderly populations. Interviews may be preferred over questionnaires due to the personal component and the opportunity for clarification of instructions. Questionnaires should contain only necessary data, and have enlarged, black print on a bright-colored paper, with graduations for ease of turning pages.
A diversity of research designs with quantitative and qualitative data should be used to fully explore gerontology. Scientists must find ways to build on previous research conclusions, such as synthesis and meta-analysis. Additionally, more studies should be replicated to lend support to what we think we know.
- Birren, J.E., Cunningham, W.R. Research on the psychology of aging. In J.E. Birren, W.K. Schaie (Eds.), Handbook of the psychology of aging. New York: Van Nostrana Reinhold, 1985.
- Brimmer, P.F. Funding for conceptual model building, journal of Gerontological Nursing 1987; 13(9):24-27.
- Brooke, V. Initiating clinical research in a longterm care setting. Journal of Gerontological Nursing 1988; 14(11):16-20.
- Burns, N., Grove, S.K. The practice of nursing research: Conduct, critique, and utilization. Philadelphia: WB Saunders, 1987.
- Cameron, C. The ethnographic approach: Characteristics and uses in gerontological nursing. Journal of Gerontological Nursing 1990; 16(9):5-7.
- Carper, B.A. Fundamental patterns of knowing in nursing. ANS Adv Nurs Sci 1978; 1(1):13-23.
- Chinn, P.L., Kramer, M.K. Nursing's patterns of knowing. In RL. Chinn, M.K. Kramer (Eds.), Theory and nursing: A systematic approach. St Louis: CV Mosby, 1991, pp. 1-17.
- Duffy, L.M., Wyble, S.J., Wilson, B., Miles, S.H. Obtaining geriatric patient consent. Journal of Gerontological Nursing 1989; 15(1):21-24.
- Engle, V.F., Graney, M.J. Meta-analysis for the refinement of gerontological nursing research and theory. Journal of Gerontological Nursing 1990; 16(9):12-15.
- Gueldner, S.H. The relationship between proposed motion and human field motion in elderly individuals living in nursing homes. In U. Malinski (Ed.), Explorations on Martha Rogers' science of unitary human beings. New York: Appleton-Century-Crofts, 1986, pp. 161-171.
- Gueldner, S.H., Hanner, M.B. Methodological issues related to gerontological nursing research. Nurs Res 1989; 38(3): 183-185.
- Hinshaw, A.S., Heinrich, J., Bloch, D. Evolving clinical nursing research priorities: A national endeavor. J Prof Nurs 1988; 4(6):398, 458-459.
- Kirchhoff, K.T. Design of questionnaires and structured interviews. In M.A. Mateo, K.T. Kirchhoff. Conducting and using nursing research in the clinical setting. Baltimore: Williams and Wilkins, 1991, pp. 135-143.
- Kovach, C.R., Knapp, T.R. Age, cohort, and time-period confounds in aging research. Journal of Gerontological Nursing 1989; 15(3):11-15.
- O'Leary, P.A., McGill, J.S., Jones, K.E., Paul, P.B. Gerontological research: Is it useful for nursing practice? Journal of Gerontological Nursing 1990; 16(5):28-32.
- Partridge, R. Learning styles: A review of selected models. J Nurs Educ 1983; 22(6):243-248.
- Pfeiffer, E. A short portable mental status questionnaire for the assessment of organic brain deficit in elderly patients. American Geriatrics Society 1975; 23(10):433-441.
- Polit, D.F., Hungler, B.P. Nursing research: Principles and methods, 3rd ed. Philadelphia: JB Lippincott, 1987.
- Robb, S. The elderly in the United States: Numbers, proportions, health status, and use of health services. In A.G. Yurick, B.E. Spier, S.S. Robb, N.J. Ebert (Eds.), The aged person and the nursing process. Norwalk, CT: Appleton-Century-Crofts, 1984, pp. 33-64.
- Ryden, M.B. Morale and perceived control in the institutionalized elderly. Nurs Res 1984; 33(3):130-136.
- Wells, T.J. Nursing model and research compatibility: Concerns and possibilities. Journal of Gerontological Nursing 1987; 13(9):20-23.
- Whall, A.L. Research approaches to development of gerontological nursing theory [Editorial]. Journal of Gerontological Nursing 1990; 16(9):3-4.
- Williams, M.E., Retchin, S.M. Clinical geriatric research: Still in adolescence. J Am Geriatr Soc 1984; 32:851-857.