A modern analgesics pain ‘pyramid’


  • R. B. Raffa PhD,

    Corresponding author
    1. Department of Pharmaceutical Sciences, Temple University School of Pharmacy, Philadelphia, PA, USA
    • Correspondence: R. B. Raffa, PhD, Department of Pharmaceutical Sciences, Temple University School of Pharmacy, 3307 N. Broad Street, Philadelphia PA 19140, USA. Tel.: +1 215 707 4976; fax: +1 239 597 7566; e-mail: robert.raffa@temple.edu

    Search for more papers by this author
  • J. V. Pergolizzi Jr MD

    1. Department of Medicine, Johns Hopkins University School of Medicine, Baltimore, MD, USA
    2. Department of Anesthesiology, Georgetown University School of Medicine, Washington, DC, USA
    Search for more papers by this author


What is known and objective

In an effort to provide guidance for the use of analgesics for pain management – while at the same time acknowledging the professional, patient and regulatory–legal concerns about the use of strong opioids – the World Health Organization (WHO) in 1986 suggested a conservative stepwise approach. In addition to the use of non-pharmacologic measures, the WHO recommended that pharmacotherapy be initiated using a non-opioid analgesic first and then progress through ‘weak’ opioids or analgesic combinations to ‘strong’ opioids if, and only if, needed. This approach gave a rationale, and a justification if necessary, for the use of opioids. This stepwise approach became widely known as the WHO analgesic ‘ladder’.


Since the initial WHO guidance, there have been significant changes in the understanding of pain. It is increasingly considered a physiological process that merits and deserves independent treatment. In addition, more analgesic options are available now than in 1986.

What is new and conclusion

Because of the evolving understanding of the physiology of pain and better approaches to its management, we suggest that more modern best practice is an analgesic ‘pyramid’.

What is Known and Objective

Prior to 1986, there were few or no published guidelines regarding the use or choice or type of analgesic drug for patients who required more than management of acute pain. Indeed, pain management as a goal independent of the cure of the underlying medical condition was in its infancy and had not yet fully emerged as a healthcare discipline.[1] In addition, the fear of addiction was, and continues to be, a strong countering force.[2] Regulatory–legal constraints and sanctions and fears of prosecution established a real or perceived impediment to the use of opioid analgesics for patients who had non-cancer pain or for long-term use in any patient, even for end-of-life patients.[3, 4] Morphine and other opioids were legal but severely restricted in many countries and legal but not culturally well accepted in other countries.[5, 6] In some developing countries, nearly 70% of patients with cancer in severe pain received no opioid.[7] The introduction of the 1986 WHO stepwise approach (a 3-step ‘ladder’) offered a simple and conservative treatment algorithm for pain control. It was based on the patients’ ‘level’ of pain (e.g. mild, moderate or severe). Physicians could begin to treat a pain patient using non-opioid analgesics, such as one of the non-steroidal anti-inflammatory drugs (NSAIDs) or paracetamol (acetaminophen). If the pain persisted or got worse due to a deteriorating disease progression, a so-called weak opioid analgesic could be introduced alone or in combination with a non-opioid. If the pain persisted or got worse due to deteriorating disease progression, a so-called strong opioid analgesic could be introduced. This stepwise approach provided everyone concerned the sense that the use of an opioid would be avoided unless absolutely necessary. The fear of addiction was abated and the fear of legal prosecution or professional censure was assuaged. An added value of the pain ladder was that it could be deployed anywhere in the world, even in countries that had few pain specialists. Following the WHO algorithm resulted in adequate pain relief for an adequate amount of the time for an adequate number of patients.[8]

However, hallmark concepts about pain physiology have emerged,[9] and it is now recognized that there are different types of pain (e.g. nociceptive, neuropathic, etc.) and that pain is multimodal (due to varied and multiple underlying physiological processes) – and therefore, it is less effective to treat pain according to its ‘level’ than it is to treat it by matching the analgesic's mechanism of action with the pain's underlying (patho)physiology.[10-12] Thus, an update of the 1986 ‘ladder’ seems appropriate.


An important corollary of selecting an analgesic agent that matches the mechanism of analgesic action with the physiology of the pain is that proper matching achieves the same effect with lower doses, better outcome and fewer adverse effects. It can also be opioid sparing. For example, severe pain associated with metastasized prostate cancer in the spine that is inadequately controlled by morphine or other opioid might be better relieved by the addition of a ‘weak’ analgesic such as paracetamol than by higher, and more problematic, doses of the opioid. The more modern approach gives better flexibility and a more holistic approach to the management of pain that alleviates some of the fears associated with the use of opioids, such as that addiction is the inevitable outcome of the use of opioids. The approach stresses the appropriate use of opioids – neither under- nor overutilization – both of which are inappropriate uses.

Why modernize the ladder?[13] The WHO ladder served as an extremely valuable conceptual and practical guide for the pharmacologic management of pain, particularly for the non-specialist. Strict adherence to the ladder successfully increased the level of adequate treatment of pain to 60%, but allowance for informed deviations added an additional 20% to the success rate, to more than 80%.[14] An update would also reflect modern practice and modern therapeutic options. First, for example, the use of pain pharmacotherapy today attempts to balance the relief of pain with quality of life. For example, although opioids are effective analgesics for many (but not all) types of pain, they have well-known and sometimes treatment-limiting adverse effects. Increasing the dose to achieve greater pain relief can lead to diminishing returns and decline in quality of life. Thus, options must be provided such that this is not the inevitable outcome. Second, there are now more classes of analgesics than there were in 1986. Instead of only NSAIDs, paracetamol and opioids, there are now multimodal agents (e.g. tramadol and tapentadol), drugs for migraine (e.g. the ‘triptans’), gabapentin, pregabalin, adjuvant medications and others. There is also radiotherapy, peripheral nerve blocks and other non-pharmacologic interventions. Third, the multimodal approach places more emphasis on combinations of mechanistic approaches rather than on lockstep increase in dose along the same mechanistic pathway.

What is New and Conclusion

A generalized pain pyramid is shown in Fig. 1. Some similarities and distinctions to the WHO ladder are highlighted by the following three scenarios:

Figure 1.

A generalized representation of an analgesics pain ‘pyramid’. Such a ‘pyramid’, as opposed to the 1986 ‘ladder’, reflects the advances made in understanding the physiology of pain and provides clinical flexibility and a patient-individualized care strategy to achieve better pain management. The different trajectories indicated are just three of many possibilities. Of particular note are the options for analgesics switching on the same step (e.g. switch to a different NSAID, or opioid rotation), or movement down steps (e.g. immediate post-operative transition to maintenance therapy), and adjuvant and non-pharmacologic interventions (e.g. creams, injections, and nerve blocks). This also broadens the applicability of the representation beyond that of the original ‘ladder’ (viz. to cancer pain) to all types of pains.

  • Patient #1 presents with mild acute pain. There is no indication of neoplasm or any progressive disease. In addition to non-pharmacologic measures (e.g. application of heat or cold as appropriate, physical therapy, etc.), the patient would be started on Step 1 of non-opioid pharmacotherapy. There are currently many options. The individual choices are not the subject of the present topic. But let us say that in this case, an NSAID or paracetamol is selected. A reasonable choice. For this example, let us say that it is observed that pain relief is not complete. One option would be to increase the dose. This might be perfectly appropriate if efficacy was limited by insufficient dose, if the dose remains below safety concerns, or if the mechanism of pain is matched to the mechanism of analgesic. For the sake of this example, let us say that the pain has an inflammatory component. The mechanism of paracetamol does not match this mechanism of pain (paracetamol is not anti-inflammatory), and therefore, used alone, it will not fully relieve this type of pain, irrespective of the dose used. Increasing the dose will only increase the possibility of adverse effects. Movement to Step 2 should be considered, not because the pain is more severe, but because the mechanism is different. The appropriate match of mechanisms of pain and analgesia would provide better pain control at safer doses. A ‘pyramid’ provides alternatives. In the figure, this is represented by sequential changes along the same level of the pyramid. Such a clinical option is not suggested by a ‘ladder’. Step 2 options also might include the use of a combination product, or a pharmacologic plus non-pharmacologic combination, or a multimodal analgesic such as tramadol or tapentadol. Providing multiple options for the opportunity to stay on Step 2 reduces the concern that progression to opioids is an inevitable consequence of initiating therapy.
  • Patient #2 presents with pain caused by a progressive, untreatable cancer. This is the classic scenario for the WHO ladder. The level of pain and disease are linked, and the increasing level of pain (‘mild’ → ‘moderate’ → ‘moderately severe’ → ‘severe’) is treated with the matching strength of analgesic (non-opioid → ‘weak’ opioid → ‘strong’ opioid). Neither aspect of this approach (the classification of pain by ‘level’ or the classification of analgesic by ‘strength’) is now considered best practice. In the pyramid, the same patient would progress upward on steps defined by a match of the type of pain and the mechanism of action of the analgesic agent. Clearly, the steps would involve some of the same agents as the ladder and ultimately involve the use of morphine or other high-efficacy opioid agonist, but they should be used and dosed based on the mechanistic considerations, not ‘strength’. Blind use based on ‘strength’ will only lead to ineffective increase in dose and consequential increase in the rate of adverse events, excess tolerance development, opioid-induced hyperalgesia[15] and other counterproductive phenomena. The mechanistic approach allows for consideration of the counterintuitive, yet beneficial, addition of a ‘weak’ analgesic to a ‘strong’ analgesic as in the example of metastatic prostate cancer in the spine discussed above. And the top level no longer needs to be restricted to opioids, and it can include interventional pain control (e.g. peripheral nerve blocks).
  • Patient #3 demonstrates perhaps the biggest difference from the view extant when the WHO ladder was created based on the level of pain and the more modern view based on the mechanism of pain. If the mechanism of pain is not matched with the analgesics typically listed on lower steps, initiating therapy with those drugs will be ineffective, unduly delay adequate pain relief and likely set in motion physiological changes that result in peripheral or central sensitization, chronification of pain and transition to more treatment-retractable pains (e.g. neuropathic). In this situation, it is better to start with the appropriate Step 4 agent. As Step 4 is not based on the level of pain, but rather the mechanism, this need not be an opioid. The figure also highlights an important feature. Pain management should be considered a dynamic process continuously assessed, individualized, time varying and flexible. If pain treatment is initiated using the properly matched analgesic, it is quite possible that the once-believed inevitable ascending progression is avoided and descending to the next lower step can be contemplated.


Since the original WHO 3-step pain ladder was introduced more than 25 years ago, there have been significant advances in the basic science understanding of pain anatomy and physiology, the number and diversity of available therapeutic agents for pain, and the societal mores and attitudes towards pain and pain treatment.[16] It seems apparent that these advances should be incorporated into an update and modification of the ladder, at the same time retaining its intent, simplicity and generality. It seems to us that much is gained by the seemingly small but meaningful change in representation to a pyramid. Neither the WHO pain ladder nor the pain pyramid is meant to substitute for evidence-based guidelines regarding the treatment of specific cases, which will differ, depending on the type of pain, the type of patient and the route of drug administration.[17, 18] They are intended to provide a construct for thinking about and following a pain treatment strategy, a strategy that has options and includes trajectories that are not inevitable and only unidirectional, but take into consideration the multimechanistic nature of pain, the individual patient's response – including their phenotypic absorption, distribution, metabolism, elimination (ADME) characteristics – and options for handling treatment failure of particular drugs within a pharmacologic class of drugs. Importantly, it also provides a series of decision points, where efficacy and safety of an individual analgesic can be assessed and addressed with alternative options if needed. Such an adjustment would be viewed as simply part of a natural process and would avoid continued progression along a suboptimal pathway. Such switches would not be viewed as treatment failures, but merely part of the pain management optimization process. The overall goal is a template that provides a better balance between efficacy and safety, recognizing both the need for adequate treatment and for avoiding safety and abuse issues.


Jo Ann LeQuang (LeQ Medical, Angleton, TX) assisted in preparation of the manuscript.

Conflict of Interest Statement

The authors declare that there are no conflict of interests.