The assessment of pain is widely recognised as a key component of care and the first step in the development of an effective pain management plan and in guiding decision-making. Effective pain management needs to be age appropriate, condition appropriate, and include regular assessment. Pain is a subjective, internal phenomenon and therefore can be influenced by a many internal and external factors. The biopsychosocial model of pain recognises the sensory, cognitive and social factors which contribute to the overall experience of pain.  Pain assessment, therefore, is a process where the person in pain expresses their pain experience, verbally or behaviourally and clinicians and carers observe this expression, interpret and act. Pain assessment tools are based on either the patient’s own report of their pain or the patient’s observed behaviour.  It involves initial, detailed evaluation of each type of pain, and regular reassessment of severity and response to treatment. 
People identified with palliative care needs experience pain for many reasons and assessment should include identification of underlying causes and types of pain. Broadly speaking, pain can be nociceptive and neuropathic, or both could be present.  Nociceptive pain is the most common pain experienced by people in general and is caused by direct injury to the tissues of the body. Neuropathic pain, which arises from injury to, or involvement of the nervous system itself, is a particularly challenging type of pain to assess and treat. Neuropathic pain is more common in some patient groups, although any patient can develop this type of pain. The prevalence of neuropathic pain is around 8 per cent in the general population it may be much higher in patients accessing palliative care services due to the association of neuropathic pain with conditions such as cancer, HIV and renal disease. [1,5] The prevalence of neuropathic pain ranges from 30 percent in cancer patients  to over 60 percent in patients receiving haemodialysis. 
Whilst some pain experienced by patients will be recent onset and other pain will be associated with procedures, some patients may present with chronic pain conditions.  Some non-malignant illnesses are associated with high levels of pain in the early stages of the condition.  In that case patients may develop chronic pain type responses to treatment, such as hyperalgesia or central sensitisation and may present with a more complex pain picture, requiring a more comprehensive pain assessment. [1,2,9]
Several validated pain assessment tools are available to assist with identification of pain for specific patient groups or specific pain types. Pain assessment tools can be broadly grouped into self-report tools, either unidimensional or multidimensional, and observational tools.
The self-report unidimensional tools, such as the Verbal Numeric Rating Scale, or the Visual Analogue Scale measure pain intensity only and have been widely used across multiple settings and multiple patient groups. There are visual variations for these scales available for children and for older adults.  This form of pain assessment remains the common approach in acute pain but has limitations for certain groups, such as those with cognitive impairment or non-verbal critically ill patients. [11,12]
Pain intensity or severity is an important measurement in pain assessment but the complex nature of some pain requires a multidimensional assessment tool.  These tools measure intensity as well as other components of pain such as interference. Tools developed for non-palliative care settings, for chronic non-cancer pain, or neuropathic pain have been tested in palliative cohorts. The Brief Pain Inventory (BPI) and Brief Pain Inventory-Short Form (BPI-SF) has been widely used in chronic pain and has been shown to be valid and reliable in advanced cancer,  in patients with COPD,  in patients with end stage renal disease  in older adults  and in children.  Other multidimensional tools such as the McGill Pain Questionnaire (MPQ) and its Short form [18,19] and the Paediatric Pain Questionnaire (PPQ)  were developed for chronic pain populations but have been tested in cancer pain and palliative care groups with some success. [9,16,19]
Multidimensional tools specifically designed to screen for the presence of neuropathic pain in the general population have been assessed in cancer pain,  in HIV related pain  in older people [16,22] and paediatrics.  The Leeds Assessment of Neuropathic Symptoms and Signs (LANSS), Douleur Neuropathique 4 (DN4) and painDETECT have all been tested in various patient groups with mixed results for validity and reliability.  For the paediatric population the Paediatric-modified Total Neuropathy Score (peds-mTNS) was developed for children who may have chemotherapy induced peripheral neuropathy and may prove useful in detecting neuropathic pain in children.  While some reviews support the use of these tools in complex pain, like that seen in palliative care, some caution should be taken as each tool has limitations and may not adequately discriminate neuropathic pain in patients with mixed pain profiles. [22,23]
If self-report is not possible then behavioural observational pain assessment tools might be used. A number of systematic reviews have reinforced the complexity of assessing pain in patents who unable to report their own pain due to age,  cognitive development or disability [24-26] and critical illness.  Assessment tools based on the observation of behaviours in the person experiencing pain have been validated in infants and young children, [9,24,25] cognitively impaired children,  cognitively impaired older adults  and patients in intensive care,  but few of these tools have been validated in a palliative care setting.
For children who are unable to self-report pain, proxy reporting by parents has been validated using multidimensional tools such as the BPI and the PPQ.  A number of studies have sought to validate tools for use in cognitively impaired older adults, including the Pain in Advanced Dementia (PAINAD)  and Doloplus 2.  The Abbey pain scale remains popular in both Australia and the UK yet there has been limited research into validating this tool.  There is considerable variation in the reliability of pain assessment tools and their clinical utility needs further investigation , particularly in use at the end of life.  The use of behavioural observations tools may be enhanced if those using them have some familiarity with the person in pain. 
- There is no single pain assessment tool currently in use which is regarded as ideal, and the use of any pain assessment tool should be supported by a comprehensive clinical assessment which should include:
- Pain intensity
- Temporal pattern
- Treatment and exacerbating / relieving factors
- Pain location
- Pain interference. 
- Other dimensions that are important are pain quality, affective aspects of pain, the duration of pain, pain beliefs and pain history.  Pain is a biopsychosocial process so the psychological impact and social impact of pain should also be assessed.
- Multidimensional tools are useful in assessing complex pain and their use is enhanced by providing patients with the opportunity to express their pain experience 
- For ongoing monitoring of pain intensity, self-report, using a simple visual analogue scale or numerical rating scale is appropriate for specific groups:
- In Children > 3 Wong Baker Face Scale 
- In Children > 7 VAS or VNRS 
- In older adults VNRS or Pain Thermometer. 
- Families and caregivers should be provided with a written pain management plan, which is updated upon reassessment of the pain and includes pharmacological and non-pharmacological approaches. 
- For children unable to self-report due to age, validated behavioural observation tools should be used, like the Faces Legs Activity Cry Consolability (FLACC) . For children with cognitive impairment the modified FLACC may be useful. 
- For adults with cognitive impairment, using an appropriate, validated pain assessment tool such as PAINAD or the Doloplus 2 is recommended. [16,26]
- Pain tools for use in non-verbal critically ill adults have also been reviewed, and the Critical Care Observation Tool (CCPOT) is able to be recommended for use in this population. 
- The prevalence of neuropathic pain can be high in some life-limiting illnesses and the use of a validated assessment tool may be useful in directing treatment. Research in this area is limited but the LANSS, DN4 and painDETECT can be recommended for use. 
- Cultural and linguistic differences also create significant barriers to pain assessment.  Using a pain assessment tool in the patient’s own language has the potential to improve at least one element of this complex situation.