A large, urban, Southeastern Veterans' Administration facility undertook this study in an effort to re-examine and revise their restraint policy. This descriptive study observed the frequency and nature of the application of restraints on nonpsychiatric extended care and nursing home units in relationship to a variety of factors. Such interest in releasing patients from their bonds and increasing their freedom, independence, and autonomy has been spurred by the statistic that every day in the United States, approximately one-half million people can be found tied to their beds and chairs in hospitals and nursing homes (Blakeslee, 1988).
The most frequently reported reasons in the literature for nurses' use of restraints are to prevent patients from harming themselves or others (Applebaum, 1984; Frengley, 1986; Strumpf, 1988; Warshaw, 1982; Warshaw, 1988); limit movement (Katz, 1981); maintain treatment plans (Lichentenstein, 1987); or control patients exhibiting confusion or agitation (Cohen-Mansfield, 1986; Gillick, 1982; MacLean, 1982). Problem behaviors, especially falls (Strumpf, 1988; Catchen, 1983); uncontrolled mobility; or resistance to treatment - such as attempts to remove intravenous lines or catheters - commonly precipitate the use of restraints (Lichentenstein, 1987).
Ironically, even though patient safety is cited most often as an explanation for applying restraints, several empirical studies reported that restraint use increased the chance of patient harm. Coyle (1979) and others (Kustadborder, 1983; Lund, 1985; Lynn, 1980; Walshe 1979) showed that restraint use did not eliminate the risk of injury; restraint use often exacerbated falls, as patients attempted to untie restraints (Strumpf, 1988; Feist, 1978; Innes, 193). Moreover, Cape (1985), Dube and Mitchell (1986), and Katz, Weber, and Dodge (1981) found that restraint use occasionally resulted in patient strangulation.
The use of restraints often has other nonanticipated, negative psychosocial effects in addition to physical harm. Gerdes reported that restraint use increased patients' disorganized behavior (Gerdes, 1968); Patrick (1967), Castleberry and Seither (1982), Miller (1975), and Oster (1976) concurred, noting that restraint use often increased confusion by decreasing sensory stimulation, limiting communication, and increasing disorientation or regression. Finally, Miller (1975) and Oster (1976) described that reciprocal social isolation occurs when patients are restrained, with staff viewing the restrained patient as dangerous or incompetent.
The few published studies on the incidence /prevalence of restraint use with nonpsychiatric populations have examined hospital or nursing home patients. The frequency of restraint use with extended care patients has not been reported. Frequently, these studies have looked at small samples, medical rather than nursing diagnoses, and retrospective chart data rather than direct clinical observations. Most studies have failed to describe the type of restraints used with patients. They have examined neither the relationship between nursing staffing patterns and restraint use, nor restraint frequency on evening or night shifts or weekends.
Therefore, the investigators embarked upon the current study to determine the total number of patients placed in restraints, the type of restraints used, nursing rationales for the application of restraints, and duration of patient time in restraints. In addition, data concerning the relationship between the use of restraints and patient age, nursing and medical diagnoses, nursing category of care, and prescribed psychotropic agents were gathered. Finally, the relationship between the use of restraints and patient-to-staff ratios was examined.
To meet these study objectives, an observational, descriptive method was used. All 173 patients on two extended care and two nursing home units were observed directly four times a day on three separate days. This observational period spanned five days and included a total of 2,076 observations. Restraints were defined as mechanical devices such as mittens, vests, pelvic holders, belts, sheets, or ties applied to the patient's body or wheelchair. Chemical restraints, geriatric chairs, and siderails were not included in this definition.
In addition, each patient's "Nursing Problem List," "Weekly/Monthly Summary Note," and "Interdisciplinary Care Plan" were reviewed to note nurses' recordings related to the need for the application or removal of restraints. If no rationale was recorded, the head nurse was asked why a particular patient was restrained. Staff were not told the purpose of the observations or chart audits. Finally, all of the agency's "Incident Reports" for the six months prior to the study were reviewed to determine the number and nature of patient falls and whether or not patients were restrained at the time of the fall.
Setting and Sample
As Table 1 illustrates, of the total 173 patients, 36 were on Extended Care Unit A, 23 on Extended Care Unit B, and 57 patients each were on Nursing Home Units A and B. "Extended care" within this facility is defined as nonacute care, with 60 to 120 hospital days being the average patient stay. The average length of stay for patients on the two extended care units in this study was 103 days (about 3 months). As would be expected, the average stay for nursing home patients was much longer, with the mean being 659 days (about 13/4 year) for Nursing Home Unit A and 797 days (about two years) for Nursing Home Unit B.
Sample and Setting Demographic Characteristics
In the sample, 48 (27.7%) were black and 125 (72.3%) white; there were no other minorities. Because the setting was a veterans' facility, there was a preponderance of males in the sample (n = 167 males and 6 females). Patients' average age was 72.5, with patients on the extended care units being somewhat younger (mean age, 72) than their nursing home counterparts (mean age, 73; Table 1).
To collect and record data, the investigators designed a "Restraint Observation Data Sheet" (RODS). This sheet allowed them to record direct clinical observation data, as well as demographic (gender, race, and age) and nursing data (number of days in the facility and unit, medical and nursing diagnoses, and nursing category of care).
One day before commencing actual data collection, the investigators performed an interrater reliability check of their observations. Making independent observations and recordings on one of the four study units, the researchers showed 99% agreement on the total number of patients restrained (one investigator missed observing one patient in bed whose restraint was covered by a sheet), and 97% agreement on the type of restraint used (one observer missed a wrist restraint and another observer a vest restraint).
Actual data collection began the following day, on a Friday, and was continued on Sunday and Tuesday of the next week. This schedule allowed observations to be made on weekdays as well as over the weekend, when staffing patterns were likely to change. Each investigator made observations on one of the four research units, observing and completing a RODS sheet for each patient. To minimize bias, the researchers did not make observations on their own units. Once observations had commenced, no new admissions were included in the study; ie, only the population of admitted patients on day one of the study was included for all observations.
Clinical observations were performed from 5 am to 6 am, 10 am to 11 am, 2 pm to 3 pm, and 7 pm to 8 pm. The rationale for these times was that these were pivotal points in patients' mobility status, ie, being in or getting into a wheelchair or retiring for the evening. Charts were reviewed after clinical observations were completed to record pertinent restraint data. Archival data regarding the number and nature of patient falls were collected after the observation period, by reviewing computer census and incident reports for all patients on the four units for the previous six months.
Research Purpose 1 : Frequency of Restraint Use
Of the 173 patients in the total sample, 55 (32%) were restrained on at least one occasion during the study period. The percentage of patients restrained per unit was: 25% on Extended Care Unit A; 35% on Extended Care Unit B; 28% on Nursing Home Unit A; and 39% on Nursing Home Unit B. There was no substantive difference in restraint-frequency rates between the extended care and nursing home units. One extended care and nursing home unit restrained 25% and 28% of their patients, respectively, while the others restrained 35% and 39% of their patients.
Research Purpose 2: type of Restraints Used
A total of 77 restraints were used on 55 patients. The types of restraints, in decreasing frequency, included vests (39), wrist(s) (14), wheelchair belts or ties (14), pelvic holders (5), mitten(s) (4), and ankles (1). Few patients received more than one restraint and even fewer more than two restraints. One patient with Huntington's chorea had four restraints applied. Data showed that the restrained patients were located in bed (59%), in a chair at the desk area (16%), in the hallways, day rooms, various therapies, or in transit (15%), and in a wheelchair in the patient's room (10%).
Research Purpose 3: Nursing Rationales for the Use of Restraints
Nurses offered 60 rationales for the use of restraints. The leading rationale, offered for 63% of the restrained patient group, was to prevent injury or falls. Among other rationales were: to prevent interfering with treatment (especially for patients with nasogastric tubes - 13%); to assist patients' balance, mobility, or support (10%); to prevent harm to self or others (7%); and to prevent wandering (7%). Five patients had more than one rationale given for the use of restraints.
It is important to note that over 50% of the time, the head nurse had to be asked the rationale for a patient's restraint, because this information was not found readily in the medical record. It is also of clinical interest that of the 15 patients placed on the electronic "wanderguard" system, seven also were restrained. Therefore, the wanderguard system does not function as an alternative to restraint use in all cases; however, it may help reduce restraint usage.
Archival data from "Incident Reports" indicated that a total of 86 patients (not necessarily those from the present study) from these four units had sustained falls in the previous six months. Of these 86 patients, 24 (28%) were restrained at the time of the occurrence, while 62 (72%) were not. Nineteen or 79% of the restrained patients who fell wore a vest restraint. (A determination of injury was not made as part of this study.)
Research Purpose 4: Duration of Patient Time in Restraints
The duration of restraint application was estimated by multiplying each citing of a patient in restraints by six hours (since four observations were made every 24 hours). The duration of time that patients were restrained ranged from 6 hours to 72 hours, with a mean of 18.5 hours on day one (Friday); 20.8 hours on day two (Sunday); 16.7 hours on day three (Tuesday); and 18.7 hours for all three days. Eleven patients (20%) were restrained at all times during the 72-hour observation period.
Research Purpose 5: Relationship Between Restraints and Patients' Age, Hospital Days, Diagnoses, and Psychotropic Pharmacologic Agents
* Average Age of Restrained Patients. These data indicated that the average age of restrained patients was 73.9, while that of the nonrestrained patients was 70.3. This difference was even more apparent for patients on the extended care units; there the average age of restrained patients was 76.3, while that of the nonrestrained patients was 67.5.
* Average Stay of Restrained Patients. The average hospital stay of both restrained and nonrestrained patients in the extended care units was similar; 93 days for restrained patients and 107 days for nonrestrained patients. However, among nursing home patients, the longer the patient stayed, the less likely he was to be restrained. That is, the average stay of nursing home restrained patients was 630 days (about one-and-threequarter years), while that of the nonrestrained nursing home patients was 778 days (about two years).
* Diagnoses and Category of Nursing Care of Restrained Patients. The most frequent nursing and medical diagnoses and nursing category of care were compared in restrained and nonrestrained patients (Table 2). Self-care deficit was the nursing diagnosis given most frequently to restrained patients, while impaired skin integrity rated highest in the nonrestrained group. As might be expected, the data showed a difference in the nursing diagnosis "alteration in thought processes"; while 35% of the restrained patients had this nursing diagnosis, only 19% of the nonrestrained patients did. Even so, that only 35% of the restrained population was diagnosed as having altered thought processes was an unexpected finding.
The nursing diagnosis of potential for injury was given to 44% of the restrained patients, as opposed to 21% of the nonrestrained group. This finding is consistent with the most frequently reported rationale for the use of restraints, which was to prevent falls. The most frequent general nursing diagnoses for all patients ineluded impairment or deficit (actual or potential) in skin integrity, cal mobility, self-care, and nutrition; 47% to 55% of the patients received these diagnoses.
Nursing and Medical Diagnoses and Nursing Care Categories of Restrained and Nonreslrained Patients*
The majority (67%) of the patients on these !four units exhibited minimal self-care independence and required extensive nursing care; 26% showed moderate independence, and only 7% of these patients were independent. When comparing levels of nursing care between restrained and nonrestrained patients, one sees that 87% of restrained patients were in the extensive nursing care category, whereas only 58% of nonrestrained patients were.
Restraint Use and Staffing Patterns
Taking medical diagnoses into account, the restrained patients exhibited a high degree of dementia; dementia was noted for 28 (51%) of the restrained group and only 36 (31%) of the nonrestrained group. There was no difference in the percentage of restrained and nonrestrained patients with the medical diagnosis of Cerebral Vascular Accident (CVA). Other differentiating diagnoses between restrained and nonrestrained groups were seizure disorder, chronic obstructive pulmonary disease (COPD), and malignancies. The higher percentage of patients with seizure disorder in the restrained group also is consistent with nursing rationales for using restraints.
* Psychotropic Medications and the Use of Kestrainis.The chart review indicated that a total of 42 (24%) patients received one or more psychotropic medications during the observation period; of these, 17 (31%) were restrained and 25 (21%) were nonrestrained. Thus, restrained patients were more likely to have psychotropic medications prescribed. The most frequently prescribed medication was haloperidol, which was given to 11 (20%) restrained and 14 (12%) nonrestrained patients.
Research Purpose 6: The Relationship Between Restraint Use and Staffing Patterns
Patient-to-staff ratios and the percentage of restrained patients on each unit for each observation day were examined (Table 3). The average patient-to-staff ratio on the nursing home units was 9.3 patients for each nursing staff person. This ratio was higher than that for the extended care units, where the mean ratio was 7.28; the overall patient-to-staff ratio for all units was 8.3.
The researchers originally believed that there might be decreased staffing over the weekend. This turned out to be the case, with the mean patient-to-staff ratio being 7.9 for Friday, 9.3 for Sunday, and 7.7 for Tuesday. However, the researchers' original intuitive hunch that the frequency of patient restraint use would be inversely related to numbers of nursing personnel (less personnel = more restraining) was not clearly substantiated. On Sunday, with only one staff person for every 9.3 patients, there was the lowest percentage of restraint use for the entire period (20%). However, the longest duration of restraint use also occurred on this day, indicating that once a patient was restrained, it was likely that the restraints would be left on for a long period.
DISCUSSION AND SUMMARY
The percentage of extended care and nursing home patients who were restrained in this research (32%) approximates that reported in two previous nursing home studies showing 25% and 30% restraint use rates (Evans, 1989; Zimmer, 1984), but is lower than that shown in Dube and Mitchell's work, 41% (Dube, 1986). Since there are no published findings for the prevalence of restraint use in extended care units, comparisons cannot be made. The restraint use rate in the present study is higher than that for acute hospital settings, for which studies have cited restraint use rates ranging from 7% (Mion, 1986) to 22% (Warshaw, 1982; Robbins, 1987).
The archetypical restrained patient in this study was older (74 versus 70 years of age), was a relative "newcomer" to the unit, exhibited altered thought processes and a high potential for injury, required extensive nursing care, and wore a vest posey on the day shift while in his room to prevent falling. Whether patients were on a nursing home or extended care unit did not affect the frequency or duration of restraint use. It is important to note that while nurses verbally described the rationale for restraining patients, frequently this reason was not charted.
Although little data exist about the types of restraint that are used (Evans, 1989), the present findings corroborate those of Morrison et al (Morrison, 1987), and Strumpf and Evans (1988), who found chest restraints most commonly used. However, Morrison et al noted that this was true only in acute settings; lap belts were most frequently used in chronic settings (Morrison, 1987). Robbins et al found that wrist restraints most frequently were applied on acute care medical units in a western veterans' facility (Robbins, 1987). Similarly, wrist restraints in this research frequently were applied in the extended care units to prevent patients from removing nasogastric tubes.
As with other restraint studies, nurses cited the most important reason for restraining patients as prevention of falls (Strumpf, 1988; Catchen, 1983; Regan, 1982; Regan, 1983). Even though nursing staff offered this rationale, a search of the records showed that approximately 28% of the patients who fell were wearing a vest posey. This confirms other reports that restraint use does not, necessarily, prevent patients from falling (Strumpf, 1988; Lichentenstein, 1987; Evans, 1989; Halpert, 1986). While Robbins et al showed that 38% of restraint incidence had no chart documentation, in the present research, the rationale for over 50% of restraint incidence was not recorded on the chart (Robbins, 1987).
It is difficult to compare findings concerning the duration of restraint use with previous research, because restraint duration often was not observed - especially in retrospective chart studies - or measures were not parallel. Usually, total number of days rather than hours during which a patient was restrained were recorded. In acute care units, Frengley and Mion demonstrated a mean restraint duration of 3.2 days out of 105 observation days (Frengley, 1986); Robbins et al found a mean duration of 3 days out of 60 days (Robbins, 1987).
In nursing home units, Strumpf and Evans reported a mean restraint duration of 45 days out of 75 observation days (Strumpf, 1988). The present study's finding of an average of 18.6 restraint hours (11%) is closer to previously published restraint durations in acute care ramer than chronic settings.
Although restrained patients in this, as in many studies, were older, Gillick et al noted that age differences between restrained and nonrestrained patients disappear when the variable of cognitive impairment is taken into account (Gillick, 1982). Robbins et al noted that cognitive impairment was the only independent predictor of restraint use among all the variables studied in their work (Robbins, 1987). Mion and colleagues agreed that cognitive deficit was highly predictive of restraint use; this research corroborates their finding (Mion, 1989).
The finding that longer hospital days were related to decreased restraint use may at first appear counterintuitive. Findings of other studies regarding the relationship between length of stay and restraint use are not clear, especially because most previous work has been done either in an acute or chronic setting and does not consider patients' cognitive impairment. It may be that over time, long-term care patients become more debilitated, reducing their activity and the need for restraints. This finding may also be related to staff getting to know patients more over time and becoming more comfortable in not restraining them. This variable requires further assessment.
These restrained patients were more likely to have psychotropic medications prescribed. Some researchers have suggested that use of these pharmacologic agents is, in itself, a type of chemical restraint (Strumpf, Mion). Although a precise relationship between medication use and restraint use cannot be defined in the present study, it is reasonable to speculate that restrained patients exhibited behaviors that also necessitated psychotropic medication use. However, this clinical issue is complicated by the fact that psychotropic medication use could increase the potential for patients falling. Therefore, further study of the behaviors that precipitate the use of psychotropic agents in restrained patients would be of interest.
While no other studies have examined the effect of shift and weekends, Frengley and Mion suggested that their reported restraint prevalence rates probably were lower because they made observations only on the day shift (Frengley, 1986). Data from this research, however, indicated that restraints most frequently were applied during the day shift, rather than during evenings or nights. This high prevalence rate may occur because more nursing staff is available on the day shift to get patients out of bed and subsequently apply restraints. Higher patient-tostaff ratios actually may result in nursing staff having to leave greater numbers of patients in bed, thus lowering activity level and the need to restrain. However, the current study was not designed to specifically examine activity level and the use of restraints. Further studies are needed to observe this aspect of care.
Directions for Future Research
Successive and replicative research with this and similar populations should be implemented. Researchers could consider doing the following:
* Use a method of direct clinical observation, on all shifts and days of the week.
* Examine the duration of restraint use in hours.
* Compare acute hospital, extended care, and nursing home units within one setting.
* Continue to investigate types of restraints used and behavior precipitating their use.
* Consider effects of staffing.
* Examine the relationship between restraint use and psychotropic medication use.
* Elaborate on staff attitudes toward use of restraints (Blakeslee, 1991; Stilwell, 1991; Tinetti, 1991).
* Pursue the relationship between restraint use and psychological and emotional consequences.
* Plan experimental studies to determine the most effective alternatives to restraint use (Powell, 1989).
IMPLICATIONS FOR NURSING
It is important to examine institutional policy and procedures to determine whether they are in accordance with current trends and research findings, and make appropriate changes. On the basis of research findings, nurses should be prepared to do the following:
* Implement changes in documentation format to include greater information on behaviors leading to the use of restraints.
* Provide staff education regarding restraint use and types of alternatives available.
* Institute a strong rehabilitative and restorative nursing care and fall prevention program.
* Redesign the physical environment to make it more conducive to a restraint-free environment.
* designate adequate resources for an effective diversion activity program.
freeing the elderly of unnecessary restraint is not only a mandated goal but also an ethical one. building on past and current research can strengthen the nurse's ability to provide more dignified care of the elderly patient and more effective alternatives to restraint use.
- Applebaum, P.S., Roth, L.H. Involuntary treatment in medicine and psychiatry. Am J Psychiatry 1984; 141:202-205.
- Blakeslee, J. Untie the elderly. Am J Nurs 1988; 88:833-834.
- Blakeslee, J.A., Goldmin, B.D., Papougenis, D., Torell, CA. Making the transition to restraint-free care. Journal of Gerontological Nursing 1991; 17(2):4-8.
- Cape, R. Avoiding restraint use. Thirteenth International Congress of Gerontology Book of Abstracts. New York: International Association of Gerontology, 1985:401.
- Castleberry, K., Seither, F. Disorientation. In C. Norris (Ed.), Concept Clarification in Nursing. Rockville, MD: Aspen Publications, Inc., 1982; 309-323.
- Catchen, H. Repeaters: Inpatient accidents among the hospitalized elderly. Gerontologist 1983; 23:273-276.
- Cohen-Mansfield, J. Billing, N. Agitated behaviors in the elderly: Part I, A conceptual review. J Am Geriatr Soc 1986; 34:711-721.
- Coyle, N. A problem-focused approach to nursing audit: Patient falls. Cancer Nurs 1979; 2:389-391.
- Dube, A., Mitchell, E. Accidental strangulation from vest restraints. JAMA 1986; 256:2725-2726.
- Evans, L.K., Strumpf, N.E. Tying down the elderly: A review of the literature on physical restraint. J Am Geriatr Soc 1989; 37(1):65-73.
- Feist, R. A survey of accidental falls in a small home for the aged. Journal of Gerontological Nursing 1978; 46):15-17.
- Frengley, J., Mion, L. Incidence of physical restraints on acute medical wards. / Am Geriatr Soc 34 1986; 565-568.
- Gerdes, L. The confused or delirious patient. Am J Nurs 1968; 68:1228-1233.
- Gillick, M.R., Serrell, N.A., Gillick, L.S. Adverse consequences of hospitalization of the elderly. Soc Sei Med 1982; 16:10331038.
- Halpert, A., Connors, J. Prevention of patient falls through perceived control and other techniques. Law, Medicine, and Health Care. 1986; 14(1):20-24.
- Innes, E. Truman, W. Evolution of patient falls. QRB 1983; 9(2):30-35.
- Katz, L., Weber, E, Dodge, P. Patient restraint and safety vests: Minimizing the hazards. Dimens Health Serv 1981; 58(5):10-11.
- Kustaborder, M. Rigney, M. Intervention for safety. Journal of Gerontological Nursing 1983; 9:159-162, 173.
- Lichentenstein, H. Stone, J. Restraints in the elderly. Paper presented at the meeting of the American Geriatrics Society, Salt Lake City, UT, 1987 (March).
- Lund, C. Sheafor, M. Is your patient about to fall? Journal of Gerontological Nursing 1985; 11(4):37-41.
- Lynn, EH. Incidents: Need they be accidents? Am J Nurs 1980; 80(6):1098-1101.
- MacLean, J., Shamian, J., Butcher, R, Parsons, R., Selcer, B., Barrett, M. Restraining the elderly agitated patient. Canadian Nurse 1982; 78(6):44-46.
- Miller, M. Iatrogenic and nursigenic effects of prolonged immobilization of the ill aged. J Am Geriatr Soc 1975; 23:360-369.
- Mion, L., Frengley, J.D., Adams, M. Nursing patients 75 years and older. Nursing Management 1986; 17(9):24-28.
- Mion, L.S., Frengley, J.D., Jakovcic, C.A., Marino, J.A. A further exploration of the use of physical restraints in hospitalized patients. JAm Geriatr Soc 1989; 37(10):949-956.
- Morrison, J., Crinklaw-Wiancko, D., King, D., Thibeault, S., Wells, D.L. Formulating a restraint use policy. J Nurs Adm 1987; 17(3):3942.
- Oster, C Sensory deprivation in geriatric patients. J Am Geriatr Soc 1976; 24(10):461463.
- Patrick, M. Care of the confused elderly. Am J Nurs 1967;67:2536-2539.
- Powell, C, Mitchell-Pederson, L., Fingerote, E., Edmond, L. Freedom from restraint: Consequences of reducing physical restraints in the management of the elderly. Can Med Assoc J 1989; 141 :561-564.
- Regan, W. Restrain as needed: Nursing judgment required. Regan Report on Nursing Law. 1982; 23(3):4.
- Regan, W. Restraints and bedfalls: Most frequent accidents. Regan Report on Nursing Law. 1983; 23(10):4.
- Robbins, L., Boyko, E., Lane, I, Cooper, D., Jahnigen, D. Binding the elderly: A prospective study of the use of mechanical restraints in an acute care hospital. J Am Geriatr Socl987; 35:290-2%.
- Stilwell, E.M. Nurses' education related to the use of restraints. Journal of Gerontological Nursing 1991; 17(2):23-26.
- Strumpf, N., Evans, L. Physical restraint of the hospitalized elderly: Perceptions of patients and nurses. Nurs Res 1988; 37(3):132137.
- Tinetti, M.E., Liu, W.L., Marotti, R.A., Ginter, S.F. Mechanical restraint use among residents of skilled nursing facilities. JAMA 1991; 265(4):468-471.
- Walshe, A., Rosen, H. A study of patient falls from bed. J Nurs Adm 1979: 9(5):31-35.
- Warshaw, G. Hospital care for elderly patients. Cenfer Reports on Advances in Research 1988; 7(3):1-8.
- Warshaw, G., Moore, J., Friedman, S., Currie, C, Kennie, D., Kane, W., et al. Functional disability in the hospitalized elderly. JAMA 1982: 248:847-850.
- Zimmer, J.G., Watson, N., Treat, A. Behavioral problems among patients in skilled musing facilities. Am J Public Health 1984; 74(10):118-1121.
Sample and Setting Demographic Characteristics
Nursing and Medical Diagnoses and Nursing Care Categories of Restrained and Nonreslrained Patients*
Restraint Use and Staffing Patterns