Is Pain Assessment Feasible as a Performance Indicator for Dutch Nursing Homes? A Cross-Sectional Approach

Instruments 

The Numeric Rating Scale (NRS) is a validated pain instrument which asks residents to rate pain intensity by number (0 = no pain; and 10 = worst pain ever) (Closs et al., 2004, Jensen et al., 1986, Jensen and McFarland, 1993, Taylor et al., 2005). The Verbal Pain Scale (VPS), a 6-point verbal pain rating scale that has been validated for use in a nursing home (Closs et al., 2004, Taylor et al., 2005), was applied when the NRS was too difficult for the resident. The REPOS is a pain observation scale consisting of 10 behavioral items. The REPOS has been validated for residents (Van Herk, Dijk van, Tibboel, Baar, Wit de, & Duivenvoorde, 2009b) who are unable to report pain themselves including residents with cognitive impairment and aphasia.

The REPOS assessment starts with a 2-minute observation period during a potentially painful moment (e.g., washing and clothing); the observer scores the ten items as present or absent as they were observed. A cutoff score of ≥3 suggests a high likelihood of pain. A high REPOS score might be the result of other emotions than pain, e.g., shame or sadness. In that case, the caregivers can give an NRS of <4. This is why the REPOS is used in combination with the NRS. The NRS represents the caregiver's opinion of the client's pain, taking circumstances into account (Van Herk et al., 2009b). In 2007 and 2008, the REPOS was implemented in several nursing homes in Rotterdam (Van Herk, Boerlage, Baar, Tibboel, de Wit, & van Dijk, 2008).

Analysis 

Nonnormally distributed data are presented as median and interquartile range (IQR). Chi-squared tests were used to determine the association between nominal data. Pain was considered to be substantial when residents rated NRS or VPS (converted to a 10-point scale) as ≥4. For the REPOS, pain was considered to be substantial for any combination of REPOS ≥3 and nurse-assessed NRS ≥4.

Results 

Pain was assessed in 201 of the 221 residents. Nineteen residents were absent, and one resident refused participation because he considered pain assessment to be nonsense. The remaining study group included 122 women (60.7%) and 79 men (39.3%) with a median age of 77 (IQR 68-84) years. One hundred forty-four residents (71.6%) provided an NRS rating, six (3.0%) a VPS rating, and the REPOS was applied in 51 (25.4%). Figure 1 shows the distribution of the pain scales used for the four different types of wards. It appears that REPOS observation was needed for all residents in the psychogeriatric ward. NRS or VPS rating was feasible for most residents of the nonpsychogeriatric wards and palliative care unit and for two-thirds of the residents on the neurotrauma ward.

View Large Image Download to PowerPoint

Figure 1 Pain scales used and residents per kind of ward. NPG = nonpsychogeriatric; PG = psychogeriatric; NT = neurotrauma; PCU = palliative care unit; NRS = numerical rating scale; VPS = verbal pain scale; REPOS = Rotterdam Elderly Pain Observational Scale.

The most frequent underlying condition was diseases of the circulatory system, i.e., in 30% of the 201 residents. Other diseases related to pain were those of the nervous system (14%), musculoskeletal system and connective tissue (13%), endocrine, nutritional and metabolic disease (6%), and neoplasm (3%).

Pain was substantial in 65 residents (32%), as rated by NRS for 52 residents, by VPS for 2, and by REPOS/nurse-assessed NRS for 11.

Median pain intensity was 6 (IQR 4-7) as rated by NRS, 5 (IQR 5-7) as rated by VPS, and 6 (IQR 5-7) as rated by REPOS with nurse-assessed NRS [4 (IQR 4-5)].

Twenty-nine residents (45%) with substantial pain received pain medication of step 1 of the World Health Organization (WHO) analgesic ladder, 8 residents (12%) received (weak) opioids (step 3 of the WHO analgesic ladder, and 3 residents (5%) received a neuroactive agent. Twenty-five residents (38%) experienced substantial pain but received no pain medication.

Six residents (4%) received opioids with NRS ≤3. Residents with NRS ≥4 were administered significantly more analgesics (chi-squared test 12.1; p = .007) than those with lower NRS. Table 1 presents the medication prescriptions in the 201 residents.

Table 1. Administered Pain Medication to Residents Without and With Substantial Pain, n (%)

		Numeric Rating Scale Pain		pValue
		0-3	4-10
	WHO step 1, analgesic:	38 (28)	29 (45)	.007
	Acetaminophen	36	27
	NSAIDs	2	2
	WHO step 2, weak opioid	2 (1)	1 (1)
	Tramadol	2	1
	WHO step 3, opioid	4 (3)	7 (11)
	Transdermal fentanyl	1	3
	Oral opioids	3	4
	Antineuropathic pain	8 (6)	3 (5)
	Gabapentin (Neurontin)	1	2
	Amitriptyline (Tryptizol)	4	—
	Carbamazepine (Tegretol)	2	1
	Pregabalin (Lyrica)	1	—
	No pain medication	84 (62)	25 (38)
	Total	136 (100)	65 (100)

Discussion 

Self-report was feasible in two-thirds of the residents of this nursing home the others required a REPOS observation. Two observers approached 201 residents within 3 days. If two observers are able to collect the information within 3 days it would be feasible for the caregivers to do so within 1 week.

Pain was substantial in one-third of the residents. Although residents with substantial pain received analgesics more often, 38% of them did not at all, which suggests that pain treatment is not yet sufficient. The latter percentage is similar to those reported in the literature, ranging from 20% to 54% (Achterberg et al., 2007, Boerlage et al., 2007, Hadjistavropoulos et al., 2007, Herr and Titler, 2009, Hutt et al., 2006, Reynolds et al., 2008, Won et al., 2004).

The consequences of persistent pain or its inadequate treatment in the elderly are important. The increased risk of functional impairment, falls, slow rehabilitation, cognitive impairment, mood changes, decreased socialization, and sleep and appetite disturbances lead not only to a decreased quality of life but also to an increase in health care cost (AGS Panel, 2002, Panel, 2009). The results of the present report suggest that pain remains a relevant problem in nursing homes in The Netherlands. Regrettably, pain is not yet included in the set of performance indicators for nursing homes implemented in The Netherlands in 2008.

There is every reason to believe that adding pain assessment would stimulate the attention to pain. The question is whether it would improve pain treatment as well. It seems that improvement cannot be achieved by assessment only. It is necessary to combine assessment with either a treatment decision-tree or an individualized standing order so that nursing staff can effectively intervene when pain requires treatment (Leone et al., 2009, Van Herk et al., 2009b). It is important that pain is a regular theme during medical rounds. Physicians must look at the results of pain assessment and ask the caregivers if an intervention was effective. Complicated pain problems should be discussed within a multidisciplinary team. Such a team should preferably include at least a physician, a psychologist, a physiotherapist, and a nurse (AGS Panel, 2009, Swafford et al., 2009).

To guarantee sufficient knowledge, caregivers as well as physicians should follow training on pain assessment and pain treatment on a regular basis. A regular basis of training assures that new personnel receive the same training as sitting personnel, knowledge level about pain stays up to date, and attention to pain is stimulated (Swafford et al., 2009).

The present study shows that pain assessment is a feasible performance indicator. Quality of pain treatment would be available on the Internet for future clients, management of nursing homes, health care insurers, and the Health care inspectorate, which might stimulate the development of a best-practice treatment model. Pain assessment combined with a pain treatment algorithm makes a good combination for the improvement of pain treatment in a nursing home.