The Prevalence and Pharmacological Management of Pain Among Older Home Care Clients

By Charlemaigne Victorino, Dr. Colleen Maxwell, Dr. David Hogan

Background

Effective pain management among older persons continues to present a challenge for health care providers. Nearly 25 percent of those aged 65 years and over have chronic pain. The prevalence of pain is of particular concern for seniors in longterm care settings with estimates among nursing home residents reported to be as high as 60 to 80 percent. The appropriate assessment and management of pain among older persons is often further complicated by the co-existence of cognitive impairment.

Prevalence estimates of disabling cognitive impairment are reported to be as high as 15 percent among those 65 years of age and over, with rates of impairment increasing with age. Estimates are even higher (around 70%) among those in long-term care institutions. The presence of cognitive impairment represents an important clinical consideration in the area of pain management since it may interfere with the patient’s ability to report pain and may also increase the likelihood for under- or inappropriate treatment and potential adverse drug events.

Pharmacologic treatment is the most common method for providing pain relief among older persons. Pain medications include opioids (e.g. morphine) and non-opioid analgesics such as acetaminophen (e.g. Tylenol®) and non-steroidal anti-inflammatory drugs or NSAIDs (e.g. Advil®, Celebrex®).11,12 Other drugs that have been developed for purposes other than pain relief, called adjuvant drugs, are sometimes used to control pain. Adjuvant medications include antidepressants, anticonvulsants, and corticosteroids.

The classes of analgesics appropriate for the relief of different levels of pain have been outlined in the World Health Organization’s (WHO) 3-Step Analgesic Ladder. The WHO’s Three-Step Analgesic Ladder is based on the following steps:

Other adjuvant drugs may be used in combination with the analgesics at any of the three steps to improve pain relief and to treat symptoms that may produce or intensify pain. The Three-Step Analgesic Ladder has been found to be useful in providing pain relief for nearly 90 percent of cancer patients.

The increasing prevalence of multiple comorbid conditions and medication use (polypharmacy) with age presents additional barriers to the effective management of pain among seniors. Older persons are more vulnerable to potential harmful side effects from analgesics and drug-drug and drug-disease interactions. Differences in the levels of education among care providers also add to the difficulties faced by older patients in obtaining adequate pain control. For example, it has been found that the rate of untreated pain varies across nursing homes and this may be partially due to differences among care providers in their knowledge of appropriate pain management specific to older persons.

Despite the high prevalence of pain among seniors and the availability of guidelines for pain management such as the WHO Three-Step Analgesic Ladder, previous research suggests that pain is poorly assessed and under-treated among older persons and thus represents a particularly important quality indicator in continuing care settings. For example, studies have demonstrated significant under-treatment of pain among institutionalized elderly cancer patients. Similarly, a large study of frail older patients living in the community found a high prevalence of daily pain yet low rates of treatment among those with pain. In this study, older patients and those with cognitive impairment were at greatest risk for undertreatment.

While previous studies have focused primarily on residents in long-term care facilities, relatively less is known about the potential under-treatment of pain among older home care clients, particularly from a Canadian perspective. This study examines the prevalence of pain and factors associated with appropriate pain management among a cohort of home care clients aged 65+ years residing in Southern Alberta.

Methods

Sample

The data used in the present analyses were collected as part of a comprehensive longitudinal investigation of the determinants and consequences of medication adherence among urban and rural home care clients residing in two health regions (Calgary and Chinook Health Regions) in Southern Alberta during April 2000 - May 2001. In each region, two nurses working with their respective home care programs were hired and trained to administer the Minimum Data Set for Home Care (MDS-HC) according to a standardized 2- day training protocol.

Subjects were recruited from random samples stratified by urban versus rural residence. The final sample consisted of 330 home care clients (165 urban and 165 rural). Rural clients represented those living on a farm, acreage or in a village or town with a population less than 10,000 and residing greater than 35km from a major urban centre. Only home care clients currently receiving services under the jurisdiction of their respective health authority were eligible for participation. Other selection criteria included age 65 years or older and provision of informed consent from either the subject or a legal guardian.

Measures

The MDS-HC items examined in this study included measures of clients' socio-demographic, physical and mental health status, ability to communicate, pain frequency and severity, and receipt of pain medication (including type of analgesic). Also included were the following index measures previously developed and validated for use with MDS instruments based on subsets of items in the MDS-HC with moderate to high inter-assessor reliability estimates: the MDS Cognitive Performance Scale (MDS-CPS), the MDS ADL self-performance hierarchy scale, and the MDS Depression Rating Scale (MDS-DRS). The MDS-CPS, a hierarchical index used to rate cognitive status, ranges from 0-6 with higher values reflecting more severe cognitive impairment. The MDS-ADL self-performance hierarchy scale is based on four ADL items (personal hygiene, toileting, locomotion and eating) and ranges from 0-6 with higher values representing greater ADL dependence. The MDS-Depression Rating Scale, derived from mood and behavioural items, ranges from 0-14 with a cut-point of 3 indicating the presence of mild to moderate depression.

Pain was defined as disruptive or intense daily pain and was based on assessment items obtained from the MDS-HC. Specifically, clients having daily pain and experiencing severe or excruciating pain or pain that disrupts usual activities were included in the definition.

Details regarding clients’ medication use (prescription and over-the-counter drugs) were recorded from drug containers/bottle labels during the in-home medication inventory/pill count by study nurses. Analgesics were classified using the Anatomical Therapeutic Chemical (ATC) classification system and were grouped based on the American Geriatrics Society clinical practice guideline 10 and the Three-Step Analgesic Ladder of the World Health Organization. The analgesic groups included acetaminophen, NSAIDs, weak opioids (e.g. codeine, oxycocet), and strong opioids (e.g. morphine). Other adjuvant medications such as antidepressants, anticonvulsants and corticosteroids were not considered in this study because the database did not provide details regarding the indication(s) for these drugs. For example, it was unclear from the data available whether an antidepressant had been prescribed for depression, chronic pain, both, or some other indication.

Analysis

Descriptive and multivariate logistic regression analyses were used to estimate the prevalence of pain and potential under-treatment as well as the relative importance of clients’ age, sex, clinical, functional and cognitive status, and ability to communicate as potential correlates of under-treatment.

Results

Characteristics of the home care sample are summarized in Table 1. The mean age of the study sample was 83 years and most of the clients were women (78.5%). Approximately 72 percent had 3 or more chronic health conditions and slightly over half were identified as having a communication problem (e.g., hearing difficulty, problem making self-understood or understanding others). Mild to moderate depressive symptoms were present among 16 percent of the sample and approximately 23 percent were cognitively impaired. Relatively few clients exhibited impaired functional (10.6%) status. Disruptive or intense daily pain was experienced by 35.5 percent of clients.

Clients with disruptive or intense daily pain were significantly more likely to be women, to have 5 or more comorbid conditions and to exhibit mild to moderate depressive symptoms (Table 1). The prevalence estimates of disruptive or intense daily pain according to clients’ demographic, functional and depression characteristics are presented in Figures 1 and 2. Disruptive or intense daily pain was significantly more common among women (38.2%) than men (25.4%) and among clients with mild/moderate depressive symptoms (49.1%) relative to those without depressive symptoms (32.9%).

Among home care clients reporting disruptive or intense daily pain, nearly 18 percent were not receiving an analgesic medication (e.g., acetaminophen, an NSAID or an opioid: Figure 3). Among those receiving 1+ analgesics, NSAIDs were the most commonly used agents (53.9%), followed by acetaminophen (36.7%), weak opioids (19.6%) and strong opioids (8.6%). Over half of this group (52.1%) were receiving more than one analgesic medication. Approximately 60 percent of clients with frequent or intense pain reported that their current medications did not adequately control pain (not reported in figures or tables).

Tables and Figures are available in the print edition of STRIDE.

The practice patterns for pain treatment did not vary significantly by sex or age (Table 2). However, clients with a communication disorder were significantly less likely to receive one or more analgesic medications for their pain (75.4%) compared with those with no problems in communication (90.4%). Although not statistically significant, clients with cognitive impairment (CPS score ³2) were also less likely to receive analgesic medications for their pain (69.6%) than those without impairment (85.1%).

Table 3 presents adjusted odds ratios for not receiving analgesic medications among clients with disruptive or intense daily pain. After adjusting for age, sex and cognitive status, clients with communication impairments remained at significantly higher risk for not receiving analgesics (OR 3.75 95% CI 1.15-12.2). In other words, clients with a communication disorder were nearly 4 times more likely than those without any communication difficulties to not receive an analgesic. Adjusting for other factors, clients aged 85+ years were significantly more likely to receive analgesics (OR 0.22 95% CI 0.05-0.92) compared with those aged 65-74.

Discussion

The potential under-treatment of pain among older persons has personal and economic consequences. On the one hand, failure to adequately treat pain may result in a poor quality of life for affected individuals and their families. This study showed that home care clients suffering disruptive or intense daily pain were significantly more likely to exhibit mild to moderate depressive symptoms. However, due to the cross-sectional nature of our data, it is also possible that depressed clients may have been more likely to report pain. Economically, the under-treatment of pain may result in increased costs due to hospital readmissions associated with uncontrolled pain, and the use of additional and possibly expensive pain relief therapies.

Results from this study showed a relatively high prevalence of disruptive or intense daily pain (35.5%) among a sample of older home care clients, which is similar to the prevalence estimates of daily pain found in a larger study of community-living elderly patients. Although the majority of home care clients in this study with disruptive or intense daily pain received an analgesic, nearly 18 percent did not receive acetaminophen, an NSAID, or an opioid. In addition, nearly 60 percent of those with frequent or intense pain and receiving pain treatment reported that their current medications were not adequately controlling their pain.

The most common pain medications used in our home care sample were non-opioid analgesics (acetaminophen and NSAIDs), while weak and strong opioids were used less frequently (19.6% and 8.6%, respectively). This is surprising since the focus of this study was on clients with intense or disruptive pain, and thus a greater percentage of clients should probably have received at least a weak opioid, as suggested by the WHO Three-Step Analgesic Ladder. This may explain why a majority of the clients in our study receiving pain treatment reported that their current medications were not adequately controlling their pain.

Our findings suggest a possible reluctance on the part of some healthcare professionals in prescribing opioids to their patients and/or the reluctance of patients to use opioids to treat their pain. Concerns that the use of opioids will inevitably result in addiction, sedation, thought impairment or other intolerable side effects have been suggested as reasons for the relatively low of use of opioids for pain relief compared with non-opioids, especially among those with moderate to severe pain. However, as emphasized in current treatment guidelines, opioid analgesics can be effective and safe, even among seniors, if used appropriately. Although not assessed in the current study, the presence of medication intolerance or adverse events could also explain the lower opioid analgesic use among our older client sample.

This study of home care clients found that the prevalence of disruptive or intense daily pain was higher among those with no cognitive impairment than those with some impairment. Several studies involving residents in long-term care facilities have also shown that the reporting of pain among older persons decreases with increasing cognitive impairment or inability to communicate.

Although non-cognitively impaired residents may actually have a higher prevalence of pain (e.g., due to a higher prevalence of disabling and painful musculoskeletal conditions), other researchers have argued that the above findings do not necessarily indicate that residents with cognitive impairment or communication problems are less sensitive or at less risk of pain compared to cognitively intact seniors. Many researchers suggest the need for better strategies and tools to assess the pain experienced by the cognitively impaired.

Contrary to studies that have found an association between older age groups and undertreatment of pain, this study did not produce support for this association. In fact, older clients (especially those aged 85+ years) were significantly more likely than those aged 65- 74 to receive analgesic medications for their pain. The absence of age and sex bias in treatment patterns is a positive finding and may reflect unique care practices/interventions for pain assessment and management in these two Canadian home care settings. The lower use of analgesic medications among the youngest age group may also reflect the influence of other factors not examined in the current study (e.g., duration of pain, use of adjuvant medications). Our findings for the different age groups should also be interpreted with caution given the smaller cell sizes for those aged 65-74.

Our results suggest that clients with some cognitive impairment were more likely to not receive an analgesic for their pain compared with those who were not impaired. This finding was not statistically significant. However, the relatively small number of impaired clients in our sample may have reduced our power to detect a statistically significant association for cognitive impairment. Clients unable to effectively communicate were at a significantly higher risk for not receiving an analgesic medication. Therefore, findings from this study show some evidence of potential under-treatment of pain among vulnerable older populations, specifically among those with communication problems.

Conclusion

Disruptive or intense daily pain was prevalent among our older-aged home care sample, especially among women, and was positively associated with depressive symptoms. Although pharmacologic treatment rates were high, clients with communication problems were significantly more likely to not receive appropriate pain treatment. Among those receiving treatment, a large number indicated that current treatment efforts were inadequate. These data support the need for further education and research efforts to improve the systematic assessment and appropriate management of pain among vulnerable older persons residing in both community and institutional settings.

Recommendations

Education plays an important part in pain management. Guidelines for the pharmacological treatment of pain are useful, but they are only effective if care providers have the required attitudes, knowledge and skills to implement the recommendations. Dealing with patient concerns about dependency and addiction are also essential. Strategies to improve education about pain of both care providers and patients have been shown to be beneficial as has supporting the professional development of providers. The high prevalence of pain among older persons in community and institutional settings is a serious public health issue. Organizations providing seniors care should take a more active role in ensuring that their residents are receiving effective pain therapy.

Acknowledgements

We wish to acknowledge Ms. Shelly Vik for her significant statistical and administrative contributions to the Alberta RAI-HC study. We are also most grateful to the four home care nurses (Michelle Copeland and Sue Couchman from the Calgary Region and Bonnie Matson and Trudy Harbidge from the Chinook Region) for their data collection efforts and clinical input on the RAI-HC home care study. We also wish to thank the members of the RAI-HC research team (Dr. Scott Patten, Dr. Jeff Johnson, Lori Romonko-Slack and Mr. Chad Mitchell) for their helpful comments and suggestions and clinical assistance. Special thanks is also given to the 330 home care clients and their caregivers (formal and informal) who graciously gave of their time to participate in this important study. The RAI-HC study/ initiative was supported by an unrestricted grant from The Merck Company Foundation, the philanthropic arm of Merck & Co. Inc. Whitehouse Station, New Jersey, USA to the Institute of Health Economics.