Commentary on Postoperative pain in older people: a review of the literature. Journal of Clinical Nursing16, 84–97


S José Closs, Professor of Nursing Research, University of Leeds, Leeds LS2 9UT, UK. +44(0) 113 343 6773; E-mail:

Postoperative pain is, as the author asserts (Prowse 2007), an issue for older people, which is fraught with complexities. In 2005 in UK, 36% of all surgery (4·9 million operations) was on people aged 65 and over (NHS Health and Social Care Information Centre, HESonline 2005). The types of surgery varied but were commonly eye, orthopaedic, thoracic, cardiac, vascular, gastro-intestinal, urological and cancer procedures. Older people are generally at a disadvantage when it comes to having their pain managed effectively, whether the pain is acute or chronic. In the postoperative situation, for many older people, both kinds of pain are present and require intervention.

Of the four areas identified in the review (acute context; pain assessment; physiological changes and pharmacological management), all require multidisciplinary input, but assessment is the one which is key for nurses. Their 24-hour a day contact allows a unique opportunity to identify the presence, location and intensity of pain on a regular basis. Doing this well is a prerequisite for subsequent good management of pain. The literature suggests that it is not performed well. For example, NCEPOD (1999) showed that, even in UK hospitals, which had Acute Pain Services, 69% of older patients had no pain assessment charts. Of those who did, pain scores were recorded less frequently (71%) than respiratory rate (91%) or sedation level (83%). There are many reasons why pain is not assessed – some obvious, some more mysterious, but I would like to make a few comments about the necessity of remembering to assess chronic as well as acute postoperative pain; the apparently simple activity of asking patients about their pain; and issues related to assessing pain in the presence of cognitive impairment.

Acute and chronic pain

Many older people have underlying chronic pain. Reports vary, but a large study in Scotland showed that 60% of those aged above 75years had some kind of persistent pain (Elliott et al. 1999). The incidence of persistent pain increases with age, associated with conditions such as osteoarthritis, vascular insufficiency and peripheral neuropathies. Any assessments of pain, including those undertaken postoperatively should take this into account. For example, back pain may be exacerbated, especially after lying on a theatre table or trolley for an extended period of time. So the acute pain caused by the operation may be one of several pains, which require different interventions to make the patient more comfortable.

Asking older patients about their pain

The attitudinal barriers, which inhibit the reporting of pain by older people are clearly outlined in Prowse's paper (2007). This is a group which is much more reluctant than most to report their pain voluntarily. From this fact we can infer a need to be proactive in our assessment of pain. The frequently witnessed question from a nurse behind a drug trolley ‘do you want anything for pain?’ almost always elicits the ‘no thank you, I'm all right dear’ type of response, regardless of the level of pain. Specific questions are more successful; asking ‘have you got any pain?’ and ‘are you hurting anywhere?’ demand that the patients describe how they are feeling and indicate where they are feeling it, rather than giving polite and empty responses. This is particularly important when the acute pain from surgery is accompanied by one or more kinds of chronic pain.

The language used needs to be culturally appropriate, too. In a study of 417 orthopaedic patients 16% scored their pain as zero on verbal and numeric rating scales, but when asked to describe what they felt, they used words such as ‘ache’‘sore’ and ‘stabbing’ (Closs & Briggs 2002). More than one approach to assessing pain may be needed since many older people simply tend not to use the word ‘pain’ in the same way as health care professionals. They may respond more readily to synonyms such as ‘soreness’‘discomfort’, ‘aching’ or ‘hurting’.

Pain assessment and cognitive impairment

The author has clearly presented the problem of assessing pain in older people who are unable to communicate their need for pain relief because of cognitive decline. This is a large group of people and empirical studies have shown that, while intensity may successfully be assessed using simple intensity scales in those with moderate impairment (e.g. Chibnall & Tait 2001, Closs et al. 2004), those who have severe impairments are a very different group.

There has been a sudden upsurge in interest over the past 5–10 years in the problem of assessing pain in people who have substantial cognitive impairment. This interest can be illustrated through three recent reviews on the subject, following years when the subject was mostly ignored (Stolee et al. 2005, Herr et al. 2006, Zwakhalen et al. 2006). There have been at least 15 different behavioural assessment scales developed, mostly for use in long-term settings. For the most part these use similar indicators to one another, including facial expressions, verbalizations, non-verbal vocalizations, body movements, mood changes and changes in activity patterns or routines. Some are very lengthy, including 60+ items (the PACSLAC, Fuchs-Lacelle & Hadjistavropoulos 2004) while some are rather more brief and therefore more suitable for clinical use [e.g. the Checklist of Non-verbal Pain Indicators (CNPI); Feldt 2000]. The CNPI is the only instrument of this type which has been specifically developed for use with acute pain. It scores the presence or absence of six different behaviours on movement and at rest, producing a range of scores from 0 to 12, where 12 suggests that the presence of pain is highly likely. However, it requires more psychometric testing and greater certainty concerning its validity before it can be widely introduced into practice.

The conditions of dementia, delirium and depression present significant issues for older postoperative patients. They are rarely differentially diagnosed and treated and require a genuinely multidisciplinary approach to management. Dementia should be diagnosed only by trained staff, using formal test instruments such as the Mini-mental state examination (MMSE, Folstein et al. 1975). Confusional states may be because of sepsis, dehydration, overhydration, abnormal blood urea and electrolyte levels or hypoxia, all of which are not uncommon following surgery. There is an important role in the management of dementia, delirium and depression for medical practitioners of old age psychiatry, who at present are underrepresented in acute settings.


There are many other issues of concern for older people after surgery. However, improving pain assessment is a good first step in the process of effective pain management. A thoughtful approach to questioning older people about pain location and intensity, together with the use of appropriate scales such as the verbal rating scale or numerical rating scale, could provide an inexpensive and rapid improvement. More psychometric testing and validation of behavioural pain assessment instruments is much needed, so that those who are most neglected at present, i.e. older people with cognitive impairment, can have their pain more easily identified and therefore controlled. This review of postoperative pain is a welcome addition to the literature, and is important in keeping the problems of older people in hospital in our consciousness.