LITERATURE REVIEW. Abstract Within the dementia care environment, it is my experience that it is often unclear to nursing staff as to appropriate assessment and management of pain for clients in the palliative stage of their illness. Although nurses have their experience to guide their practice, as the health professional most involved with the client at the end of life, there is a need for review of current assessment tools and management strategies to ensure the care given is evidence based and best practice.
Aim The aim of this review is to examine what is the most appropriate method of pain assessment and management when working with clients with cognitive impairment. Methods Searches were made of several databases Cinahl Embase Medline BNI Psych For articles published between 1990 and 2006 using the keywords, pain assessment, pain management, elderly, dementia and palliative care. Findings Most evidence in relation to this area of practice is anecdotal demonstrating a need for further research.
Evidence presented in this review shows encouraging results in regard to the development of assessment tools and that there are clear standards of process for management of pain at the end of life.
The results also show that nurses are concerned and are making attempts to be proactive in this area of care despite many barriers. Conclusion Good pain control requires both sensitivity to the clients needs on every level and the competence to meet them. Assessment and management of pain for individuals with a dementia is still very hit and miss, with evidence of both very good and very poor practice being demonstrated.
Nurse education is still an unmet need within this area of practice. And that overall clients are experiencing unnecessary pain on a regular basis which could have been prevented. Introduction Pain Pain can be a common experience for many older adults and has negative effects on their health, functioning and general well being Pain presents an enormous challenge for most nurses within their daily working lives. This is due to the individual nature of pain; a client’s experience of pain is affected by their life experience compounded with the procedure or illness.
Macaffrey identified the individual nature of pain when he said, “Pain is what the experiencing person says it is and exists whenever he says it does”(1979, Cited in Epps 2001). Pain can be present in the older adult for many reasons. Chronic pain conditions may include arthritis, gout and peripheral vascular disease, acute conditions such as cancer; cardio vascular disease and surgery are also prevalent causes of pain. An elderly client may present with more than one type of pain in different locations with varying causes making the nurses job all the more challenging.
Dementia. The word dementia comes from Latin and translates as “out of ones mind”(Epps 2001). Although this gives a broad definition of the symptoms, dementia is actually a disease caused by permanent brain changes and loss of neurons. These changes affect the client in several ways. The impairment of cortical functions affects “memory, orientation, comprehension, calculation, learning capacity, language and judgement”(WHO 1993). Various dementias exist the most common being Alzheimer’s, which accounts for 65% of UK diagnosis’s (Stevens et al 2001).
There are other dementias such as vascular, lewy body and picks. All of the dementias impair basic functioning and the ability to communicate also the client’s ability to interpret the feelings of pain and recall them to communicate to the nurse. This impairment potentially results with individuals unable to express pain and report it appropriately. This in turn may result in inadequate assessment and management as has been shown in studies (Bernabei Et al; 1998. Horgas & Tsai; 1998).
One study reported that 60% of nursing home residents with a physical diagnosis known to cause pain had not received any pain relief in the previous month (Feldt et al 1998). The American geriatric society (1998) believe that clients with dementia are at high risk for unidentified and under treated pain. As with all older adults, clients with dementia present with chronic conditions such as arthritis and acute pain experienced in the aging and the end of life process.
Moss (2002) gives evidence that most elderly clients who move into long-term care will die in an institution either a nursing home or a hospital many of whom will have dementia. She states that 91% have a strong co morbid condition likely to cause pain. “The control of pain is the key to all other care”(Raiman 1998). This is especially true within old age psychiatry as if a client is in pain it may be impossible to recognise a pure diagnosis of agitation or depression and intervene accordingly when there is the possibility that these symptoms may be reactive due to inadequate pain relief.
The key to therapeutic success is how well are the symptoms being relieved. Untreated pain in the client with dementia can delay healing, disturb sleep and daily activity, reduce quality of life and prolong hospitalisation (Horgas 2003). Appropriate assessment and management of pain should be paramount to nurses working in the elderly care sector. Search methods A search of the Cinahl, Medline, Embase, BNI, psycINFO databases were made using the following key words, pain assessment, pain management, elderly, dementia and palliative care.
The reference lists of identified papers were also searched for relevant articles. The use of many databases was deemed necessary due to the specialism the enquiry relates to as research in dementia is still relatively new. Inclusion criteria are as follows: Articles must relate to palliative care and/or pain relief in the elderly that have reference to impaired cognition within the main body of the text and not just articles solely referencing dementia in order to gain a wider breath of material.
Articles must not be specific to a particular medical condition such as breast cancer as the recommendations will be specific to condition and not transferable to an elderly psychiatric unit. Articles must have been published between 1990 and 2006. The review was not limited to a study design type as most papers identified are qualitative and give anecdotal evidence. Anecdotal evidence may not be seen as the most reliable (Parahoo 1997) but there are reasons why within this area of research most results fit into this category.
Atkinson et al (1999) suggest that palliative care and adequate pain relief does not attract much funding from the pharmaceutical companies therefore unattractive to researchers, also as suggested by Ray et al (1998) ethical committee accreditation may be difficult to obtain particularly with issues as consent and suitability to participate in controlled trials as they may last longer than the client remains alive giving inaccurate results. Another reason maybe that the cognitively impaired are often excluded from research studies, as they are unable to give adequate feedback to fulfil the researchers needs for the inclusion of data.
The search returned nine articles that fully met the inclusion criteria these have been included in the review. Findings The findings will be divided into two sections pain assessment and pain management. All papers bar one fell in to the naturalistic paradigm of research in that the researchers explored the nurse’s feelings and thoughts during semi- structured interviews about assessment and management of pain and recorded their findings in a descriptive manner.
The paper set in the positivist paradigm was a correlation design that examined the relationship between two variables in this instance pain and agitation and generated results of a quantative nature. In all papers the sample was purposive in that clients were selected because they experience the phenomena under investigation in this instance, elderly institutionalised clients. All studies identified within the papers had been ethically approved. Pain assessment. A barrier to effective pain management is a failure to assess pain. Pain assessment is invaluable; it sets a base line against which the intervention can be compared.
There are several methods of pain assessment identified within the literature. This review explores the findings drawn from the research. This article will draw reference to self-report scales as they still hold their place within pain assessment as well as those methods that concentrate on adults that are unable to communicate their pain. Pain can be assessed by means of self reporting and research by Davie et al (2004), Kovach et al (2000), Panke (2003) and Zwackhalen et al (2006) identify this as the “gold standard” in pain assessment.
There are several self-report scales available to assess pain in the elderly. Most self report tools focus on pain intensity these include The visual analogue scale (VAS) which asks a client to select a number between 1 and 10 and the verbal rating scale (VRS) where you ask the client to verbalise their pain on a scale between no pain and worst pain possible. Molony (2005) found the VRS the most popular scale to use when questioning nursing staff; the main drawback of the tool that she identified was that clients learnt to anticipate the nurse’s questions. nd therefore the test may lose objectivity. There is some conflict within the literature in that Zwackhalen believes that these scales hold little benefit within dementia following her research as she identifies that these scales requires the capacity to understand the task and also the ability to communicate pain therefore generating inaccurate generated scores this view is supported by Hayes (1995) and Ferrell (1995) whereas Davies generates evidence that 80% of clients in midstage dementia were able to use these scales effectively.
It is because of these contrasts that it would be beneficial to explore behavioural tools and other methods as a means of supporting self-report tools in order to gain comprehensive assessment. Panke (2003) believes that non-verbal cues are their own means of self-report and should be regarded as such by patients that can’t communicate.
The most commonly referenced tool within the literature is The Checklist of Nonverbal Pain Indicators (CNPI) by Feldt (2000), research on this tool by Horgas (2003) and Davie (2004) have deemed this to be an effective tool to use within dementia care. There are other tools identified within the literature but due to the frequency that it was referenced and studies that included it within their framework the critique of literature in regard to this tool seemed most relevant. The tool is designed as an observational tool for use by nursing staff.
It was designed to assess postoperative clients. It asks nurses to observe for certain behaviours and score a point of one if they are present. There is no clear definition attached to this tool as to how the score correlates to pain intensity, the papers that reviewed this tool recommended that it be used in conjunction with a self reporting tool but also if there were any behavioural indicators of pain scored present further investigation and treatment should be actioned.
Both of these researchers conducted their studies in community settings which could have biased the results as the tool is designed for acute pain not chronic pain and you can make the assumption that clients were not postoperative also the samples used were small there by not possibly encompassing the full range of cognition and varying physical illness. Also this tool is subjective to the nurse’s viewpoint hence the researchers results are biased towards the nurse’s viewpoint and not the client’s experience. Epps (2001) believe that nurses often under report pain of clients with dementia.
It is assumed as the researchers indicated the use of another tool that this a limitation of this tool that it cannot be used as a stand-alone assessment. It may be worthwhile to repeat these studies in an impatient dementia environment to test the validity of the results. Behavioural indicators of pain as outlined by Evans (2002) in addition to those in Feldts tool are aggression, loss of appetite and withdrawal Kovach (2003) further expands on this by advising staff to look for changes in mobility and respiration.
Evans and Kovachs work highlights that Feldt tool is not comprehensive, as it does not highlight what people are trying to communicate with their ongoing behaviours and that it can only be used at that snapshot in time, this factor is something that was not identified by the authors. Also how often this tool should be utilised is not explored would it be beneficial to use it daily or just when the nursing staff notice change. Panke (2003) believes you should consider all cue verbal and non-verbal and that continual reassessment of these cues is vital.
Molony (2005) asked nursing staff to document the frequency of assessment. The highest frequency was found to be on admission, as determined by medical condition and those patients receiving “narcotics”. This concurred with 95% of participants. This study was comprehensive as it had 160 participants covering a cross section of job roles within the nursing sphere. Panke (2003) points out that behavioural cures are not always reliable, as moaning and grimacing are symptoms that occur with delirium as a common factor in the client in the end stage of life and that this may require assessment for both causes.
This is supported by Ferrell (1995). Cohen Mansfield et al also concluded through their study that assessment is made further difficult as the client group expressed discomfit Constipation, emotional distress, cold and hunger in the same ways such as calling out and increased agitation. Cohen Mansfield study was extremely limited as it was restricted to one nursing home with a very small sample all of which had a moderate dementia and had been placed there for over two years. Most literature reviewed did not account for other possible causes of pain like symptoms.
This is both positive and negative as the papers remain focused on the topic of pain assessment but the detraction is that it does not encourage the reader to view the client’s situation holistically. Only Kovach (1999) within my included literature highlighted the importance of the nurse client relationship in interpreting none verbal cues and that familiarity with the client develops the nurse’s intuition of changes in behaviour. Kovach cited another study by Parke (1998) within her research that advocated that assessment methods require nurses to know their patient intimately in order to recognise behaviours which signal presence of pain.
The ability to communicate and relate to your clients is important as often the nurse is the main caregiver and clients need to feel that we will respond to them and make them feel safe at a time when the situation is alien and fearful. It would be beneficial to further explore the importance of the skills needed by the nurse in this area of care. All papers identified a lack of nurse knowledge and competence as a barrier in the assessment of pain and many suggested the need for action to improve this also nurse attitude was cited.
This was mainly centred on nurse’s unfamiliarity with the assessment tools. This is identified by Davie (2004). ) Who when gaining feedback on the research conducted, the nursing staff felt that tools could be too time consuming and difficult to interpret. 99% of respondents who participated in Molony’s study (2003) felt that education was something that should be ongoing and be presented in the form of workshops, seminars and peer teaching by the bedside in order to update practice.
The main barrier identified of staff not receiving appropriate training was the need for more nurses to staff units so others could attend training. This is in conflict with the NMC code of conduct that states it is the nurses own responsibility to update her practice and therefore ward staffing should not be an issue although many trusts have policies to support learning and development. Evans (2002) reviewed the literature and other research within her study, she found that educational initiatives that address clinical skills in pain and symptom management were required.
Moss (2002) in a study of four hundred nursing homes in the united states that 29% of staff did not feel confident to adequately assess pain although 55% agreed pain was a problem for clients with dementia. This would leave 45% to believe that there is no problems with this issue hence highlighting further the need for nurse education, as almost half the staff in four hundred nursing homes did not recognise the problem. This figure is astounding.
Kovach (1999) as part of her study evaluated the education programme currently being piloted to address the discomfit needs of clients with moderate to severe dementia. She found that staff using the protocol for assessment for which they had been trained resulted in increased awareness of resident’s discomfit and improved assessments. The findings of this study could have been strengthened by the use of a control group of nurses already competent in the assessments to provide a conclusive baseline that nurses should attain and a larger sample size.
Also the recognition of variables such as physical diagnosis would have been beneficial to assess if they impacted on the findings. Glaring omissions in all research critiqued was that of involving family in the decision making process and also the gaining of information from families in order to ascertain normal behaviours and responses to pain, this could be for many reasons. Is this due to the nurse being perceived as the expert or are families at this distressing time not considered to be able to give eliable information. This requires further exploration as to the benefit of relatives input in a clients care. Also omitted was the use of do not escalate or do not resuscitate order and what they men to the client receiving palliative care and in the age of the expert patient many making advance directives and how this dictates the care we can give as nurses, do we have the right to override these statements if we deem it to be in the clients clinical best interests and where does this feature in the assessment process.
This requires further research. The use of documentation and the value of the nurse-to-nurse handover is not explored within the literature as a means to aid assessment both of which are commonplace in a nurses daily routine, which leads me to believe that both these activities happened but were not documented by the researchers. Pain management Horgas and Tsai (1998) investigated prescription administration of analgesic medication of clients with cognitive impairment.
The research question they posed was that cognitively impaired nursing home residents might not be prescribed adequate analgesia. The researchers concluded that clients with dementia were being prescribed and administered significantly less analgesic medication both in number and in dosage than clients without cognitive impairment it was also reported that clients who were more disorientated or withdrawn were prescribed significantly less analgesia. This correlation study drew from a large sample of 339 residents from four homes showing the study to be both reliable and valid.
Moss’s study provides anecdotal evidence to support that nursing staff felt that pain in client with dementia is as aggressively treated as pain in other clients whilst Panke (2003) identified that physician were loathe to prescribe for any thing that wasn’t cancer. These conflicts may be explained as the United States and the United Kingdom have different protocols for appropriate treatment disseminated from government and allied agencies and moss’s study was in the UK and the others in the U. S.
Kovach (2000) reports that the treatment strategies used by nurses are poorly understood. The outcome of her interviews with nursing staff was that nurses often felt that psychotropic drugs were being inappropriately used to treat the behavioural symptoms of pain and that analgesics commonly underused. Both the observers and the workers have identified this as a problem yet it still continues to happen as Swakhalen’s 2006 study shows. It is because of this that the need to review best management is vital.
Recent research by the Joseph Rowntree foundation indicates that these problems could be overcome by the use of the NICE guidelines for analgesia (2004) these recommend regular administration and treatment adjusted from one step to the next according to increasing or decreasing pain severity, history of analgesic response and any side effects. This approach also calls into question the use of as required medication and its prevalence. The literature concurs that regular scheduled pain relief is most effective in managing symptoms. The foundation also advocates the use of non-pharmacological intervention for Pain management.
The literature mainly identifies methods of drug administration and drug type most appropriate to the client group. Management of pain is one of the most important goals when caring for clients at the end of life. Horgas (2003) identified the goal of pain management is to maximise function and improve quality of life. Kovach (2000) found that nursing staff prioritised comfort over issues such as side effects or toxicity. Although some staff expressed concerns about sedation and falling in mobile clients but this becomes less of an issue with clients in the end stage of life, as they are normally bed bound.
The issue of constipation was also raised by Panke (2003) as a common side effect of analgesics that was likely to cause pain and recommended that clients should be prescribed laxatives as a matter of course. Horgas (2003) identified the considerations for administering analgesics in the elderly. It is noted that aging has an affect on the effect of the drug on the client and also the concentration of the active drug is affected. Elderly clients are at higher risk of side effects due to decline in metabolism and elimination.
Gradual titration is recommended to over come this. A Diamorphine as opposed to morphine is commonly the drug of choice for subcutaneous administration due to greater solubility. Kovach (2000) found that nurses felt that narcotics were underused but appreciated the value of them to maintain comfort and dignity. She supports this with several anecdotal accounts from staff. Nurses only commonly used morphine for severe pain and to promote comfort during dying. The literature concurs that longer acting opioid medication such as Fentanyl should not be rescribed for the dying client because of the slow onset of pain control and it being difficult to titrate quickly. Kovach (2000) identified that fentanyl was more commonly used than morphine in the relief of clients pain. It is recommended that fentanyl is administered via a transdermal patch as this can overcome many reasons for non-compliance in the demented client such as resistiveness; lack of understanding or forgetting to take medication also it is considered helpful for those clients that are unable to swallow oral preparations.
If fentanyl has previously been administered and deemed not effective this will need to be supplemented with diamorphine via a syringe driver as recommended by many researchers within the review. Only half of the participants in Kovach’s study were able to identify appropriate dose escalation, most identified that there was a need to increase dosage but unable to identify a systematic process within which to work. Does this mean that clients are only getting the appropriate treatment 50% of the time due to a lack of staff education?
Not all pain is opiate sensitive such as pain on movement that’s breakthrough. This is more likely to be effectively managed by the use of a non-steroidal anti-inflammatory drug (NSAID). This group can cause gastric irritation and the use of a gastric protective drug is advised. Panke (2003) identified that nurse were proactive in advocating supplementary pain relief of this nature. The literature cited the transcutaneous electric nerve stimulation (TENS) as an appropriate method of pain relief, but no evidence is available to state how effective this method actually is.
All literature that cites the TENS also states it should not be used for first line pain relief. Conventional medication may not always be affective and nurses should have an awareness of complementary therapies and relaxation. Nurses in Kovachs study were not questioned about non-pharmacological measure but common themes drawn from the interviews were that massage, one to one intervention, repositioning and music therapy were thought to be effective.
None of the literature mentions the involvement of other agencies in promoting comfort such as the tissue viability team or the music therapists, the onus is put on the nurse to provide these things which may be outside her sphere of competence also lacking within the research is the use of equipment such as airflow mattresses and relaxing fragrance atomisers or fans to assist with those that can’t maintain body temperature. Although the literature is comprehensive on chemical pain relief, it is often not holistic as to assess the client’s environment.
There are articles on alternatives to medication that were generated during the literature search such as one by Pearce (2004) advocating that the approach be high touch and include things such as hand activity and soothing touch, but sadly articles such as this that make complete sense to the reader are not evidence based and therefore inadmissible to be critiqued in this review. Actual research in this area is desperately needed to guide nurse’s practice to create a holistic approach. It should be noted that techniques such as relaxation require higher levels of functioning and may only be successful if the client can participate.
Restlessness and delirium are common close to death. This may be due to opioid toxicity, a recommendation of haloperidol is recommended for drug toxicity. Panke (2003) is the only author to identify this. I have commonly seen it myself in practice that a low dose antipsycotic is very effective. Haloperidol also can assist with nausea and vomiting. The ethical issue as to whether analgesia will hasten death is identified within the literature as a concern for nurses and relatives. With nurses interviewed by moss (2002) asking what is permissible?
The American nursing association in 1991 stated “nurses should not hesitate to use full and effective doses of pain medication for the proper management of pain in the dying patient. The increasing titration of pain medication to achieve adequate symptom control, even at the expense of life is ethically justified”. None of the literature asks what is the patient’s quality of life and if we are, why are we attempting to extend this, also could unrelieved pain hasten death due to the stress level it places on the client. These are all area that needs further research.
No mention is made in the research as to the impact of hospital policy on the treatment of these clients and whether there are set standards dictated from trust level. Conclusion There is a clear need to address this issue as the proportion of people with dementia is likely to increase along with the increasing proportion of elderly persons in the population as due to more effective health care people are living longer. Although pain assessment and treatment is complicated with the client with dementia. Many clients pain can be relieved by the nurse’s efforts to understand each individual and their unique way of communicating pain.
Effective assessment and management requires “ listening to and believing in reports of pain as well as the knowledge and skill in detecting pain among those unable to convey its presence verbally” (Horgas & Tsai 1998). All management should be individualised and tailored to the client. The most important step in improving pain control in clients with dementia as identified by this review is for nurses to be proactive in the assessment of pain. Pain assessment is the important first step in effective pain management.
This review has identified that individuals with dementia can answer simple questions about the presence of pain and pain assessment in older adults should always start with self-reported pain. Nurses need to be aware of all possible sources of information when assessing clients with dementia and other conditions that impair communication and take into account non verbal behavioural cues such aggressiveness and increased vocalisation. This review also identified the need for increased frequency of assessment and that it is performed at a relevant time as clients may not be able to recall past experience of pain due to their impairment.
Pain is a significant problem for older adults and has the potential to negatively impact on all areas of their lives. Pain management that includes both pharmacological and non-pharmacological measures was considered to be effective. The need for appropriate nurse education has been highlighted throughout the review and that most nurses felt that they did not have sufficient knowledge to care for these clients, this is something that needs urgent address in order to provide a good quality of service.
With the advancement of technology and medication, pain is not an experience that should be commonplace for any individual receiving medical care whether it is in hospital or in the community and with the effective use of strategies we as nurses can really improve the quality of our client’s lives. Areas highlighted for further research from this review are nurse education and ethical issues regarding pain relief and non-pharmacological treatments. This review found many omissions in the research but this could be due to the fact that certain activities are commonplace and the expectation is on the reader to assume these activities happen.