Review of Literature on the Effectiveness of Physical Restraints on Individuals with Alzheimer’s Disease

Physical restraints in nursing homes have received much debate in recent years. While many nursing homes continue to use restraints, it seems that researchers (Capezuti et al., 1996) tend to believe that restraints actually cause more accidents than prevent them.

A study done by Phillips, Spry, Sloane, and Hawes (2000) discusses the use of physical restraints in Alzheimer’s Special Care units (SCU) located in nursing homes. Its purpose was to assess whether or not SCU residents would be less likely to be physically restrained than regular nursing home residents because of more specialized treatment. Participants in this study included 71,000 nursing home residents in four states (Kansas, Maine, Mississippi, and South Dakota), along with 1,100 residents living in Alzheimer’s SCU. A total of 841 nursing facilities participated in this study (Phillips et al., 2000).

The primary measurement used was derived from the Minimum Data Set (MDS). The MDS is a government mandated screening and assessment form for Medicare and Medicaid certified nursing homes facilities in the United States (Frederiksen, Tariot, & De Jonghe, 1996). This document has to be completed within fourteen days of admission of a resident to the facility. In addition, it has to be updated quarterly, annually, and when there is a significant change in a resident’s status. The information collected in the MDS is used in planning the care of the resident, such as administering new forms of treatment (i.e., physical restraints). According to numerous studies, the MDS has been shown “to provide high quality data in studies where it is used by trained research nurses” (Phillips et al., 2000). Previous research also indicates that the MDS is reliable and valid (p<.001) (Frederiksen, Tariot, & De Jonghe, 1996). Results from this study (according to the MDS results) indicate that residents living in the SCUs did not differ in the amount of physical restraints used from residents living in nursing homes. In addition, it found that chemical restraints (psychotropic medication) were used more frequently in SCUs. Therefore, the original hypothesis of this study was not supported (i.e., p<.20).

Even though this study had some valuable information necessary for the literature review, it also had limitations (Phillips et al., 2000). Firstly, it barely addressed the possibility of error or researcher bias in its findings. On the other hand, it did acknowledge the possibility that results may differ based on the type of facility or type of employees at the facility. Secondly, the study does not list what information was gathered from the MDS. Since the MDS is quite long, it is expected that data used were from a section related directly to the topic. However, this information is neglectfully not listed, leaving readers to draw their own conclusion.

A study done by Weiner, Tabak, and Bergman (2003) focuses on the use of physical restraints with patients in Israel who suffer from dementia. It reviews the ethical dilemmas that can occur when using restraints on patients suffering from dementia in two types of healthcare settings: hospitals and nursing homes. In this study, the authors analyzed the different levels of education the nurses had in relation to physical restraints. A structured questionnaire was used for measurement, which consisted of three main parts: (1) demographic data, (2) a knowledge test, and (3) a list of ethical preferences related to the use of restraints. “Cronbach’s alpha was used to examine the internal reliability of each variable (α was 0.72 for knowledge of Patient’s Rights Law, 0.84 for knowledge of the Israeli Code of Ethics, and 0.68 for knowledge of the guidelines on restraints)” (Weiner, Tabak, & Bergman, 2003, p. 516). Each section of the questionnaire was judged and evaluated on its relevance by eight judges (experts in the field of nursing and internal medicine), and then it was pre-tested by 23 nurses.

The results of this study indicated that the mean level of professional education was higher in hospitals than in nursing homes (61% of nursing home aides did not have a bachelor’s degree in nursing, p<.001). Furthermore, “91.1% of the respondents from hospitals had undergone geriatric training compared with only 49% of their counterparts (p<.001)” (Weiner, Tabak, & Bergman, 2003, p. 517). As a result, nurses in hospitals tended to agree more strongly with the use of physical restraints, referring to them as “realistic and necessary” in contrast to nursing home nurses who tended to be lower educated and against their use. In addition, analysis between the benefit of the patient and that of other patients showed there was a significant correlation between the purpose of restraints (p<.001; F=55.00; df =1198), and the benefit of the patient and that of the institution (p<.001, F<21.03; df =1198) (Weiner, Tabak, & Bergman, 2003).

There were several noticeable limitations to this study which should be mentioned. To begin with, the research focused solely around the perspectives and education of the nursing staff versus the reactions of the other employees such as social workers. Another limitation of this study was that it did not address the type of dementia or how many years’ since patients had been diagnosed (Weiner, Tabak, & Bergman, 2003). This information could have been useful to future researchers, especially since studies in favor of physical restraints are found to be quite limited (Castle, 2000). Although the measurements prove that this study was reliable and valid, this is found to be questionable. Since the measurement tool used was designed by the “experts” in this study, it is difficult to determine its accuracy without repetition. In addition, the study does not discuss the possibility of biased answers from the nursing staff. For these reasons, this study cannot be considered reliable and valid without replication to prove its accuracy. However, because this study’s focus was directly alike to this paper (unlike most of the research found), I decided to incorporate it.

Another valuable resource for this paper was written by Robinson et al. (2007) which is a systematic review of the literature on the effectiveness and acceptability of non-pharmacological interventions to reduce wandering of dementia patients. As previously discussed, wandering is only one form of behavioral problem occurring with people who have dementia. Because there are so many varied non-pharmacological treatment options for patients who wander, this study compares and contrasts eleven of the best studies (11 out of 278) found dealing with this topic. Some of the options include: exercise, games, therapeutic touch, essential oils, and of course use of physical restraints. Studies chosen met criteria based on study population, outcomes, intervention, and quality.

The results of this study were inconclusive because there was no robust evidence to recommend a better intervention over the rest (Robinson et al., 2007). Music therapy and exercise were found to be the most popular forms of non-pharmacological interventions. Across the board, physical restraints were listed as “unacceptable.” However, as the authors point out, “reporting of studies was generally poor and so the quality of the conduct of the studies was uncertain” (p. 13). Personal biases and the ethical issues around the use of physical restraints are also listed as possible limitations in the accuracy of the data. Another limitation is that not all studies are completely comparable (i.e. qualitative versus quantitative). Despite these limitations, the chosen research studies were considered the best in the field, providing the most reliable and accurate information to date on this topic (Robinson et al., 2007).

The Alzheimer’s Association also published their own literature review on the topic of use of physical restraints to prevent falls and wandering. Tilly and Reed (2006) screened 109 articles and found 28 relevant enough to be reported. These articles were required to have a sample size of 10 or more residents, use an experimental or quasi-experimental design with a comparison group, and report their statistical findings. The results of these studies showed that physical restraints caused more harm (i.e., increased behavioral issues, p<.05) than the prevention of incidents (Sullivan-Marx et al., 1999).

Use of physical restraints has been found to be more common if a patient has a cognitive impairment (Sullivan-Marx et al., 1999). Two other “extensive literature reviews (Capezuti, 2004; Castle & Mor, 1998) document the negative effects of restraints on nursing homes residents, which include agitation, infections, and physical de-conditioning” (Tilly & Reed, 2006, p. 8). In addition, thirteen studies demonstrate that education and training supported by experts can reduce the use of physical restraints. The authors also list four studies (Ejaz, Jones, & Rose, 1994; Capezuti, Evans, & Strumpf, 1996; Neufeld et al., 1999; Evans, Wood, & Lamber, 2002) that supposedly prove physical restraint reduction reduces the risk of incidents among patients with AD. What the authors neglect to share with the reader is that all four of the studies listed were not specific to individuals with Alzheimer’s. Only one of the studies listed focused on the topic of dementia (Capezuti, Strumpf, Evans, & Maislin, 1999). Even though evidence broadly (for all nursing home residents) suggests the physical restraint reduction, it would be interesting to find evidence that specifically shows the decrease in incident rates. Overall, the studies reviewed were all somewhat relevant to the topic, and each has proven similar results, thus adding to the validity of the individual studies.

The nursing home reform provisions of the Omnibus Budget Reconciliation Act of 1987 (OBRA) were implemented to make improvements in the mental health care of nursing home residents. An article written by Snowden and Roy-Byrne (1998) reviews the literature revolving around the changes made by OBRA and how it affected nursing home residents. Such changes included preadmission and annual screening of residents to ensure that individuals who were diagnosed as mentally ill were not improperly admitted. In addition, “the law specifically prohibited the use of physical restraints for discipline or convenience or if they were not necessary to treat actual symptoms” (Snowden & Roy-Byrne, 2000, p. 231).

According to Snowden and Roy-Byrne (1998) the theoretical reasons given by researchers around why physical restraints are still used have all been disregarded thus far. One example could be that physical restraints would be cheaper than hiring more staff. When comparing the cost of care between nursing home residents, studies found that residents who were restrained required higher care nursing costs versus the unrestrained resident. Another theoretical reason individuals give for why physical restraints are still used involves being short staffed. Although one study found by Snowden and Roy-Byrne (1998) claimed no association between staffing ratio and the use of physical restraints, other studies of larger magnitude (Graber & Sloane, 1995; Phillips, Hawes, & Mor, 1996) have consistently reported that fewer staff had a definite impact on how frequently restraint methods were used. A few studies also have linked the prevalence of the use of restraints with a subsequent increased risk of serious falls. Factors listed that could have contributed to these injuries might include “de-conditioning from inactivity while restrained, their motor impairment from tranquilizers frequently given for the increase in agitation associated with initiation of restraints, and staff’s failure to alternate restraints and to exercise and reposition residents who are in restraints” (Snowden & Roy-Byrne, 1998, p. 230). As a result of the new regulations by OBRA, much evidence supports it has contributed to improved care and reductions in the use of chemical and physical restraints.

Some limitations found with this review involved “inappropriate use and management of restraints were common” during testing (Snowden & Roy-Byrne, 1998, p. 230). Specifically, residents were often not released from their physical restraints, repositioned and exercised as the state and federal guidelines call for, even though medical chart notes indicated that it had been done (Schnelle, Newman, & White, 1992). In addition, there were several limitations in this article for the purposes of this paper. One, the article addresses use of restraints with all nursing home residents, not just individuals who have AD. For this reason, the statistics found do not necessarily accurately represent this population. Because the passage of the nursing home reform provisions (OBRA) brought about new changes in the usage of physical restraints, it was necessary to be addressed for this paper. Also, specifics of each study reviewed were not listed, which left the reader dependent on the authors’ use of reliable and valid resources. However, most studies in this article were in agreement with one another, in addition to being chosen from the best studies available on this topic.

After researching many articles on the topic of physical restraints, it seems that the overwhelming conclusion is that physical restraints are harmful to nursing home residents, and do not particularly aid in the prevention of incidents. So one may ask…why are they still so widely used if there is no benefit to the patient or the facility?

As Robinson et al. (2007) points out, the reporting of studies has been “generally poor and so the quality of the conduct of the studies” are uncertain. For instance, the majority of the studies did not include specifics pertaining to the instruments they used, and also barely addressed possible limitations that could have occurred over the course of study. Because of the ethical issues revolving around the topic of physical restraints, it seems many are obviously in favor of restraint-free nursing facilities. Some studies (Tilly & Reed, 2006), although useful, were extremely one-sided, only recognizing the negative effects of restraints, and positive advantages to restraint reduction.

Since OBRA implemented new regulations, steps have been taken to reduce the improper usage of physical restraints. Although many studies have been implemented since 1987, very few have specifically addressed patients in nursing homes with AD. Even the literature review sponsored by the Alzheimer’s Association referenced articles that were not specific to the population. After much research, it is evident more relevant studies are needed pertaining to this topic.

Technorati Tags:


, , , , , , , , , , , , ,