Long-term and high dose opioid medicine use in the U.K.

Harvey, Jane Ellen (2018) Long-term and high dose opioid medicine use in the U.K. PhD thesis, University of Nottingham.

[img] PDF (Thesis - as examined) - Repository staff only - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (3MB)



The number of prescriptions dispensed for opioid medicines has increased in the U.K. in the last two decades and studies have shown patients are receiving opioids for longer periods than in the past. There is a lack of evidence as to the effectiveness of these drugs when used long-term, as efficacy evidence is taken from short clinical trials in populations who do not have the comorbidities commonly seen in chronic pain patients. Large observational studies of patients prescribed opioid medicines outside of clinical trials, have identified that some patients taking long-term opioids are reporting they are still in high levels of pain. There is also a concern that patients are receiving high dosages of opioid medicines without effective pain relief. However, no studies in the U.K. have looked at the proportion of patients who continue to receive opioids (for all conditions) in the long-term or at high dosages so we do not know how opioid use develops in the U.K..

Research from other countries has also identified that long-term and high dose use is linked to patient characteristics such as patient demographics and psychological and physical comorbidities. For example, a prior history of depression has been found to increase the risk of long-term and high dose use. This is of concern as this may indicate that patients may be potentially medicating the depression with the opioid. This may not be the case in the U.K. due to key factors such as the structure of the health system, so the aim of this thesis was to see if this phenomenon could potentially be occurring in the U.K.


This thesis was a retrospective observational cohort study of patients receiving opioid medicines for non-cancer conditions using data from a U.K. derived primary care database, the clinical practice research datalink (CPRD). Patients were included in the study if they received a prescription for opioid medication at any point in the year 2009 and followed, where possible to January 2015 (though data was collected for baseline variables prior to 2009). This thesis consists of a series of longitudinal analyses that attempt to define and describe the probability of long-term use and proportion of patients who receive high dosages in the U.K and seeks to understand how baseline characteristics (such as having a comorbid condition occurring before opioid use starts) affect the probability of long-term and high dose use. Competing risks methods were used to calculate the probability of continuation of opioid medicines (using death as a competing risk) and to determine long-term use. Cox regression models were used to determine whether baseline factors (such as prior antidepressant use) were associated with discontinuation of long-term use. Calculation of odds ratios were used to predict whether baseline characteristics predicted high dose use. Cox regression was also used to predict time to discontinuation of high dose use.


In the U.K., 10.58% of patients received opioid medicines to treat non-cancer pain in the year 2009. Of the non-cancer patients included in the study, 41.41% were patients who received opioids in the six months prior to 2009 and the remaining were new users of opioid medicines. In the new user opioid group, 5.75% continued to be prescribed opioids for at least one year. When including new and existing users of opioid medicines, one in thirty people in the U.K. population who started opioids for non-cancer pain in 2009 were continuously prescribed opioids for a full year. Patients who were female, received an antidepressant before opioid use started and were in the youngest age category were more likely to continue opioid use past 2 years. The probability of continuing to take opioid medicines is higher in patients who have been receiving opioids for longer and are on higher dosages; in patients who had received over 10mg OMEQ for over two years, over half of patients continued to receive prescriptions for opioid medicines for the five year period study after 2009.

In the U.K. population, one in a hundred opioid medicine users were prescribed a high dose (estimated dosage received >120mg oral morphine equivalents per day for at least 91 days) for non-cancer pain in their first 91 days of use in the year 2009. Patients receiving high dosages were likely to be receiving multiple opioid drug types and to receive preparations with multiple release profiles.

In new users of opioid medication, the odds of high dose use were increased in patients who were younger (aged 18-49 years), had 3 or more comorbidities or were in receipt of an antidepressant or benzodiazepine and/or anti-anxiety drug before or after opioid use started. In new and existing users of opioid medicines, patients who were receiving high dosages were more likely to be diagnosed or be recorded with symptoms of depression or anxiety and to have been prescribed an antidepressant or benzodiazepine and/or other anti-anxiety drug in the youngest age group compared to the older age groups.

Of the patients that were prescribed high dosages of opioid medicines for at least 91 days, 37.64% of patients did not have a second consecutive high dose quarter due to death or stopping high dose use. Almost one in five patients who had a high dose quarter continued to have a high dose for the next three years (22.98%). Older patients, patients prescribed weak opioids and/or tramadol discontinued high dose use at a faster rate than younger patients.


Both high dose and long-term use were found to be associated with a prior prescribing of antidepressants before opioid use started, suggesting that in the U.K. psychological comorbidities are associated with continued and high dose opioid use. Further work is required to measure outcomes within these groups and to understand the care that these patients have already received.

Most patients who start opioid medicines in the U.K. stop taking them in their first year of use. However, of those who continue use past two years, a large proportion continue for the full five year period. Similarly, only a small proportion of patients receive high dosages of opioid medicines but once this use is established, many patients have a high probability of continuation. Further work should be undertaken to facilitate effective review of these patients in practice.

Item Type: Thesis (University of Nottingham only) (PhD)
Supervisors: Knaggs, Roger
Chen, Li-Chia
Subjects: R Medicine > RC Internal medicine > RC 321 Neuroscience. Biological psychiatry. Neuropsychiatry
Faculties/Schools: UK Campuses > Faculty of Science > School of Pharmacy
Item ID: 52175
Depositing User: Harvey, Jane
Date Deposited: 30 Aug 2018 09:48
Last Modified: 20 Jul 2020 04:32
URI: https://eprints.nottingham.ac.uk/id/eprint/52175

Actions (Archive Staff Only)

Edit View Edit View