2
PATTERNS OF OPIOID USE IN PATIENTS WITH CANCER AND NON-CANCER PAIN AND ITS
RELEVANT CLINICAL OUTCOMES IN NON-CANCER PAIN
BY
ASWEEN ROWENA BINTI ABDULLAH SANI
A thesis submitted in fulfilment of the requirement for the degree of Doctor of Philosophy in Pharmacy Practice
Kulliyyah of Pharmacy
International Islamic University Malaysia
MAY 2020
ii
ABSTRACT
Opioids are the gold standard for the treatment of moderate to severe acute pain and cancer pain. However, opioids are increasingly prescribed to treat non-cancer pain for long-term which has become a public health concern due to increasing evidences of opioid-related harms such as abuse, misuse, and opioid overdose-related deaths associated with long-term opioid therapy. Little is known about the patterns of opioid prescribing among non-cancer pain patients and the characteristics of those on long-term opioid therapy in Malaysia.
Information is also lacking on the risks of opioid abuse/misuse among non-cancer pain patients receiving opioid treatment for their pain. This study therefore investigated individual patient level opioid prescribing patterns and the differential patterns of opioid persistence between cancer pain and non-cancer pain patients at Malaysian outpatient tertiary hospital settings over 3 follow-up years (part 1). This study also investigated the clinical outcomes and risks of opioid abuse/misuse in non-cancer patients attending pain clinics (part 2). For part 2, non-cancer patients were stratified into short-term and long-term opioid user groups. The clinical outcomes were assessed using the Brief Pain Inventory – Short Form (BPI-SF), the Medication Quantification Scale – III (MQS-III), the Short Form- 36v2 (SF-36v2) Health Survey and assessed for opioid adverse effects. The risk of opioid abuse/misuse were assessed using the Screener and Opioid Assessment for Patients with Pain - Revised (SOAPP-R). Factors influencing long-term opioid use and high risk of abuse/misuse were also analysed using multivariable logistic regression. Opioids included in this study were dihydrocodeine, oxycodone, morphine, fentanyl, and buprenorphine. In the first part of the study, a total of 922 patients with a mean age of 60 (±15.4) years who received opioids were identified. A comparative analysis indicated that compared to cancer pain patients (n = 665), non-cancer pain (n = 257) patients were prescribed relatively lower doses of opioids of <100 mg/day and for longer periods of at least 60 days of opioid days’
supply on average in a follow-up year. The differential persistence patterns of opioid use over time revealed a pattern of tapering off opioid treatment among persistent opioid users in the non-cancer pain group in contrast to the cancer pain group. In the second part, a total of 61 non-cancer pain patients were recruited where no significant differences in the clinical outcomes were found between the short-term (n=30/61) and long-term opioid users (n=31/61). Notably, 62.3% of these non-cancer pain patients were found to be at high risk of opioid abuse/misuse. Predictors of high risk of abuse/misuse included younger age (OR 0.90, 95% CI 0.86, 0.98) and higher pain interference (OR 2.17, 95% CI 1.14 – 4.13). These findings suggest that opioid prescribing practices for non-cancer pain at outpatient tertiary hospital settings in Malaysia is in accordance with opioid prescribing guidelines which recommend against high dose opioid therapy and encouraging tapering off opioid treatment. Nonetheless, the high proportion of non-cancer pain patients attending pain clinics at high risk of opioid abuse/misuse is worrisome which calls for further investigation into the risks of opioid abuse/misuse among non-cancer pain patients attending pain clinics.
iii
ﺚﺤﺒﻟا ﺔﺻﻼﺧ
ت�ﻮﻴﻓﻷا ﻲﻫ
رﺎﻴﻌﳌا ﱯﻫﺬﻟا جﻼﻌﻟ مﻻﻵا ﺔﻴﻧﺎﻃﺮﺴﻟا ﺔﻄﺳﻮﺘﳌاو
ةدﺎﳊاو . ﻊﻣو
،ﻚﻟذ ﺪﻗ ﺢﺒﺻأ جﻼﻌﻟا ﻞﻳﻮﻃ
ﺪﻣﻷا ﻦﻋ ﺎﻬﻘﻳﺮﻃ مﻻﻶﻟ ﲑﻏ ﺔﻴﻧﺎﻃﺮﺴﻟا رﺪﺼﻣ
ﻖﻠﻗ ﺔﺤﺼﻠﻟ ﺔﻣﺎﻌﻟا
. ﺰﻛﱰﺗ تﺎﺳارﺪﻟا ﺔﻘﻠﻌﺘﳌا
ﻂﻤﻨﺑ ﺎﻬﻔﺻو
تﺎﺟﺮﳐو تﺎﺳارﺪﻟا
ﺔﻳﺮﻳﺮﺴﻟا ﰲ
لوﺪﻟا ﺔﻣﺪﻘﺘﳌا . ﺎﻤﻨﻴﺑ ﰲ
�ﺰﻴﻟﺎﻣ ﻼﻓ فﺮﻌﻳ ﻻا ﻞﻴﻠﻘﻟا لﻮﺣ طﺎﳕا ﻒﺻو ﻩﺬﻫ
ﺔﻳودﻷا ﻰﺿﺮﳌ ﱂﻷا ﲑﻏ ﱐﺎﻃﺮﺴﻟا ﺪﻣﻸﻟ
ﻞﻳﻮﻄﻟا ﺮﻃﺎﳐو ءﻮﺳ ﺎﻬﻣاﺪﺨﺘﺳا .
ﻖﻘﲢ ءﺰﳉا لوﻷا اﺬﳍ ﺚﺤﺒﻟا ﻦﻣ
طﺎﳕأ ﺎﻬﻔﺻو ﻰﻠﻋ ىﻮﺘﺴﳌا يدﺮﻔﻟا
ﻰﺿﺮﳌ ﱂﻷا ﱐﺎﻃﺮﺴﻟا ﲑﻏو
ﱐﺎﻃﺮﺴﻟا ﺎ�وﺬﺧ�و
ﰲ تادﺎﻴﻌﻟا ﺔﻴﺟرﺎﳋا
ﻊﺒﺘﺘﺑ
ﲏﻣز ثﻼﺜﻟ تاﻮﻨﺳ . ﰎو ﺎﻀﻳأ ﺺﺤﻓ طﺎﳕﻷا ﺔﻴﻠﺿﺎﻔﺘﻟا ﺎﻬﻣاﺪﺨﺘﺳﻻ
ﺮﻤﺘﺴﳌا ماﺪﺨﺘﺳﺑﺎ
سﺎﻴﻘﻣ تﺎﺒﺜﻟا
ﻦﻤﻀﺘﳌاو ﺔﺛﻼﺛ
دﺎﻌﺑأ كﻼﻬﺘﺳﻻ ﺔﻳودﻷا
. ﺰﻛر ءﺰﳉا ﱐﺎﺜﻟا ﻦﻣ ﺔﺳارﺪﻟا ﻰﻠﻋ ﻰﺿﺮﻣ ﱂﻷا ﲑﻏ ﱐﺎﻃﺮﺴﻟا ﻦﻳﺬﻟاو
ﰎ
ﻢﻬﺘﻟﺎﺣا ﺎﻘﺣﻻ
ﱃا تادﺎﻴﻋ ﱂﻷا
. ﰎ ﻢﻴﻴﻘﺗ ﺞﺋﺎﺘﻨﻟا ﺔﻳﺮﻳﺮﺴﻟا ﻞﻜﻟ
ﻦﻣ ﲔﻣﺪﺨﺘﺴﳌا ﻋ
ﻰﻠ ىﺪﳌا ﲑﺼﻘﻟا ﻞﻳﻮﻄﻟاو
ﻊﻣ ﺔﻤﺋﺎﻗ دﺮﺟ ﱂﻷا ةﺮﺼﺘﺨﳌا -
جذﻮﻤﻨﻟا
،ﺮﺼﺘﺨﳌا سﺎﻴﻘﳌا
ﻲﺋاوﺪﻟا
،ﻲﻤﻜﻟا جذﻮﻤﻨﻟا
ﲑﺼﻘﻟا ﺔﺳارﺪﻟ ﺔﺤﺼﻟا
ﻢﻴﻴﻘﺗو رﺛﺎﻵا ةرﺎﻀﻟا ﺎﳍ . ﺎﻤﻛ ﰎ ﻢﻴﻴﻘﺗ ﺎﻀﻳأ ﺮﻄﺧ ﻲﻃﺎﻌﺗ ءﻮﺳو ﺎﻬﻣاﺪﺨﺘﺳا ىﺪﻟ
ءﻻﺆﻫ ﻰﺿﺮﳌا ﻦﻋ ﻖﻳﺮﻃ زﺮﻔﻟا
ﺢﻘﻨﳌا . ﰎ ﻞﻴﻠﲢ ﻞﻣاﻮﻌﻟا ﱵﻟا ﺮﺛﺆﺗ ﻋ ﻰﻠ ماﺪﺨﺘﺳا ت�ﻮﻴﻓﻷا
ةرﻮﻄﺧو ءﻮﺳ
ﺎﻬﻣاﺪﺨﺘﺳا ﻦﻋ
ﻖﻳﺮﻃ راﺪﳓﻻا
ﱵﺴﺟﻮﻠﻟا دﺪﻌﺘﻣ
تاﲑﻐﺘﳌا . تﺮﻬﻇأ ﺞﺋﺎﺘﻨﻟا ﺔﻧرﺎﻘﳌﺑﺎ ﻊﻣ
ﻰﺿﺮﻣ ﱂﻷا ﱐﺎﻃﺮﺴﻟا ﻪﻧأ
ﰎ ﻒﺻو ﻰﺿﺮﳌ ﱂﻷا ﲑﻐﻟا
ﱐﺎﻃﺮﺳ تﺎﻋﺮﺟ ﻞﻗأ
ﺎﻴﺒﺴﻧ
>
100 ﻎﻣ / مﻮﻳ ةﺪﳌو 60 مﻮﻳ ﻰﻠﻋ ﻞﻗﻷا ﰲ ﻂﺳﻮﺘﳌا ﻦﻣ
م�أ ءﺎﻄﻋا ت�ﻮﻴﻓﻷا
ﻼﺧ ل ﺔﻨﺳ ﺔﻌﺑﺎﺘﻣ . ﻒﺸﻛ ﻦﻳﺎﺒﺗ طﺎﳕﻻا ةﺮﻤﺘﺴﳌا ﺎﻬﻣاﺪﺨﺘﺳﻻ
ﻊﻣ روﺮﻣ ﺖﻗﻮﻟا ﻦﻋ ﻂﳕ ﻦﻣ جﻼﻌﻟا ﺎﺑﻬ ﻞﻜﺸﻟﺑﺎ
ﺺﻗﺎﻨﺘﳌا ﲔﺑ
ﺔﻋﻮﻤﳎ ﻰﺿﺮﻣ ﱂﻷا ﲑﻏ ﱐﺎﻃﺮﺴﻟا ﻰﻠﻋ
ﺲﻜﻌﻟا ﻊﻣ ﺔﻋﻮﻤﳎ ﻰﺿﺮﻣ ﱂﻷا ﱐﺎﻃﺮﺴﻟا .
ﰲ ءﺰﳉا
ﱐﺎﺜﻟا ،
ﰎ ﻒﻴﻇﻮﺗ 61 ﺾﻳﺮﻣ ﻦﻣ ﻰﺿﺮﻣ ﱂﻷا ﲑﻏ ﱐﺎﻃﺮﺴﻟا ﱂو
ﻢﺘﻳ ﺔﻈﺣﻼﻣ يا
قﺮﻓ ﲑﺒﻛ ﰲ ﺞﺋﺎﺘﻨﻟا ﺔﻳﺮﻳﺮﺴﻟا ﲔﺑ
ﻲﻣﺪﺨﺘﺴﻣ ت�ﻮﻴﻓﻷا
ﻰﻠﻋ ىﺪﳌا ﲑﺼﻘﻟا ﻎﻟﺎﺒﻟاو
ﻢﻫدﺪﻋ (n=30/61) ﻞﻳﻮﻄﻟا ىﺪﳌا ﻰﻠﻋ ﲔﻣﺪﺨﺘﺴﳌا ﲔﺑو
(n=31/60). ﺎﻬﻣاﺪﺨﺘﺳا ءﻮﺳ ﺐﺒﺴﺑ ﺮﻄﺨﻠﻟ ﲔﺿﺮﻌﻣ ﱐﺎﻃﺮﺴﻟا ﲑﻏ ﱂﻷا ﻰﺿﺮﻣ ﻦﻣ % 62.3 نأ ﺪﺟو . تاﺆﺒﻨﺗ
ﺮﻄﳋا ﲑﺒﻜﻟا ءﻮﺴﻟ ﺎﻬﻣاﺪﺨﺘﺳﻻا ﺖﻠﴰ
ﻟا ﻦﺴ
ﺮﻐﺻﻷا (OR 0.90, 95% CI 0.86, 0.98)
ﻞﺧاﺪﺘﻟاو ﱄﺎﻌﻟا
ﱂﻸﻟ (OR 2.17, 95% CI 1.14 – 4.13) ﻒﺻو تﺎﺳرﺎﳑ نأ ﺞﺋﺎﺘﻨﻟا ﻩﺬﻫ حﱰﻘﺗ ت�ﻮﻴﻓﻷا ﻰﺿﺮﳌ
ﱂﻷا ﲑﻏ ﱐﺎﻃﺮﺴﻟا ﰲ
تﺎﻴﻔﺸﺘﺴﳌا ﺔﻴﺟرﺎﳋا
ﰲ
�ﺰﻴﻟﺎﻣ ﻖﻓاﻮﺘﺗ ﻊﻣ تادﺎﺷرا ﺎﻬﻔﺻو
و ﱵﻟا
ﺢﺼﻨﺗ مﺪﻌﺑ لوﺎﻨﺗ ﺔﻋﺮﺟ ﺔﻴﻟﺎﻋ ﺎﻬﻨﻣ و ﻒﻗﻮﺗ جﻼﻌﻟا ﺎﻴﳚرﺪﺗ . ﻊﻣو
،ﻚﻟذ نا ﺔﺒﺴﻨﻟا ﺔﻴﻟﺎﻌﻟا ﻰﺿﺮﳌ ﱂﻷا ﲑﻏ
ﱐﺎﻃﺮﺴﻟا ﻦﻳﺬﻟاو
نودﺗﺎﺮﻳ تادﺎﻴﻋ ﱂﻷا
ﺮﻣأ ﻖﻠﻘﻣ ﺚﻴﺣ ﻢ�ا نﻮﺿﺮﻌﻣ ﺮﻄﳋ
ءﻮﺳ ماﺪﺨﺘﺳا ت�ﻮﻴﻓﻷا
. اﺬﻫو
ﺚﳛ ثﻮﺤﺒﻟا ﺔﻴﻠﺒﻘﺘﺴﳌا
ﺪﻳﺰﻤﻠﻟ ﻦﻣ ﻖﻴﻘﺤﺘﻟا ﻒﻴﺻﻮﺗو
ءﻮﺳ ﺎﳍﺎﻤﻌﺘﺳا ىﺪﻟ
ﻰﺿﺮﻣ ﱂﻷا ﲑﻏ
ﱐﺎﻃﺮﺴﻟا .
iv
APPROVAL PAGE
The thesis of Asween Rowena Abdullah Sani has been approved by the following:
_________________________________
Che Suraya Mohd Zin Supervisor
__________________________________
Zaswiza Mohd Noor Co-supervisor
__________________________________
Abdul Hadi Mohamed Co-supervisor
__________________________________
Lisa Nissen Co-supervisor
__________________________________
Aznan Md Aris Internal examiner
__________________________________
Mohd Baidi bin Bahari External examiner
__________________________________
Noorizan Abd Aziz External examiner
__________________________________
Jesni Shamsul Shaari Chairman
v
DECLARATION
I hereby declare that this thesis is the result of my own investigations, except otherwise stated. I also declare that it has not been previously or concurrently submitted as a whole for any other degrees at IIUM or other institutions.
Asween Rowena Abdullah Sani
Signature ……… Date ………..
vi
INTERNATIONAL ISLAMIC UNIVERSITY MALAYSIA DECLARATION OF COPYRIGHT AND AFFIRMATION OF
FAIR USE OF UNPUBLISHED RESEARCH
PATTERNS OF OPIOID USE IN PATIENTS WITH CANCER AND NON-CANCER PAIN AND ITS RELEVANT CLINICAL
OUTCOMES IN NON-CANCER PAIN
I declare that the copyright holder of this thesis is jointly owned by the student and IIUM.
Copyright © 2019 Asween Rowena Abdullah Sani and International Islamic University Malaysia. All rights reserved.
No part of this unpublished research may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise without prior written permission of the copyright holder except as provided below
1. Any material contained in or derived from this unpublished research may only be used by others in their writing with due acknowledgement.
2. IIUM or its library will have the right to make and transmit copies (print or electronic) for institutional and academic purpose.
3. The IIUM library will have the right to make, store in a retrieval system and supply copies of this unpublished research if requested by other universities and research libraries.
By signing this form, I acknowledged that I have read and understand the IIUM Intellectual Property Right and Commercialization Policy.
Affirmed by Asween Rowena Abdullah Sani
……… ...…...…..…………...
Signature Date
vii
This thesis is dedicated to my beloved parents, Abdullah Sani and Adeda Rahimah, for their endless love, unconditional support, and encouragement.
viii
ACKNOWLEDGEMENTS
First and foremost, all praises are due to Allah, the Most Beneficent and the Most Merciful, for His mercy and blessings which enabled me to complete this thesis.
I would like to express my heartfelt gratitude to my enthusiastic supervisor, Assoc.
Prof. Dr. Che Suraya Mohd Zin, for her guidance, patience, motivation and invaluable knowledge imparted throughout the course of this research project and in accomplishing this thesis. I would also like to express my appreciation to my co-supervisors, Assoc. Prof.
Dr. Abdul Hadi Mohamed, Assoc. Prof. Dr. Zaswiza Mohd Noor, and Prof. Lisa Nissen for their insightful suggestions, support, and encouragement during the course of this project.
I would also like to express my deepest appreciation to all staff who were involved in this project at pharmacies and pain clinics at Hospital Tengku Ampuan Afzan (HTAA), Hospital Kuala Lumpur (HKL), and Hospital Selayang for their kind support and assistance in facilitating the process of data collection and patient recruitment for this project. Many thanks to all patients who participated in this study as well.
Special acknowledgement to Dr. Michael Ling King Hwa, for his guidance and advice in the earlier days of my PhD journey in UPM, and for his reassurance to pursue this doctoral degree. I have learned the values of selflessness, high ethics, and passion for science during my brief period of working under him in UPM.
Thanks also to my dear friends in UPM for the short getaways and luncheons which uplifted me in my moments of crisis and kept me going on my path to success. Their friendship has made my PhD journey a wonderful experience. Special thanks to my husband and strong supporter, Hamzeh Al-Khatib, for his prayer, encouragement, and support in completing this thesis.
Last but not least, my greatest gratitude is towards my parents who have been a constant source of strength throughout my journey, and also to my brother, Asrul Hadi, and my sister, Asreen Ayumi, for giving me the encouragement I needed during my PhD journey. Words cannot express my appreciation for their unconditional support and especially tolerance during my lengthy periods of absence from important family events.
This journey would not have been possible without their complete understanding, patience, and unfailing belief in me.
Once again, I would like to convey my greatest appreciation for everyone that encouraged, supported, and assisted along the completion of this thesis. May this research benefit the future of pain management in this country.
ix
TABLE OF CONTENTS
Abstract ... ii
Abstract in Arabic ... ii
Approval Page ... iv
Declaration ... v
Acknowledgements ... viii
List of Tables ... xiii
List of Figures ... xvi
List of Abbreviations ... xix
Conference Presentations ... xx
List of Publications ... xx
CHAPTER ONE: INTRODUCTION ... 1
1.1 Background of the study ... 1
1.2 Pain ... 3
1.2.1 Pathophysiology of Pain ... 3
1.2.2 Acute pain versus Chronic pain ... 4
1.2.3 Biopsychosocial Model of Chronic Pain... 6
1.2.4 Chronic Non-Cancer Pain (CNCP) ... 8
1.3 Opioids as pharmacological treatment for pain ... 14
1.3.1 Opioids ... 16
1.3.2 Neurobiology of Opioids and Mechanism of Action ... 17
1.3.3 Clinical Potency of Opioids ... 18
1.3.4 Opioid Efficacy ... 18
1.3.5 Adverse Effects ... 20
1.3.6 Tolerance, Dependence, and Addiction ... 22
1.3.7 Opioids For CNCP ... 24
1.4 Problem Statement ... 27
1.5 Research Hypothesis ... 27
1.6 Aims and Objectives of the Research ... 28
1.6.1 Objectives ... 28
1.7 Significance of the Study ... 29
1.8 Thesis Structure and Outline ... 30
1.9 Ethical Approval ... 32
CHAPTER TWO: LITERATURE REVIEW ... 33
2.1 PATTERN OF OPIOID USE ... 33
2.1.1 Opioid Use Patterns in Countries with High Average Consumption of Opioids ... 33
x
2.1.2 Opioid Use Patterns in Countries with Low Average Opioid Consumption
... 36
2.1.3 Opioid Use Patterns in Malaysia ... 37
2.2 ASSESSMENT OF PAIN, HEALTH-RELATED QUALITY OF LIFE, AND RISKS OF OPIOIDS USE ... 39
2.2.1 Unidimensional Tools of Pain Assessment ... 40
2.2.2 Multidimensional Tools of Pain Assessment ... 42
2.2.3 Health-related Quality of Life (HRQoL) Assessment ... 43
2.2.4 Risk of Opioid Abuse and Misuse Assessment ... 45
2.3 MEDICATION PERSISTENCE TO OPIOIDS IN CNCP PATIENTS ... 47
2.3.1 Medication-taking Behaviour ... 47
2.3.2 Definition of Persistence ... 47
2.3.3 Measurements of Persistence ... 48
2.3.4 Persistence for Opioids ... 48
2.3.5 Factors Influencing Persistence to Opioid Therapy in Patients with CNCP ... 49
CHAPTER THREE: PATTERNS OF OPIOID USE AND THE DIFFERENTIAL OPIOID PERSISTENCE PATTERNS IN CANCER AND NON-CANCER PATIENTS OVER A 3-YEAR FOLLOW-UP PERIOD ... 51
3.1 PATTERNS OF OPIOID USE IN CANCER AND NON-CANCER PAIN OVER A 3-YEAR FOLLOW-UP PERIOD ... 51
3.1.1 Introduction ... 51
3.1.2 Aims and Objectives ... 53
3.1.3 Methods ... 54
3.1.4 Results ... 70
3.1.5 Discussion ... 83
3.1.6 Strengths and Limitations ... 95
3.1.7 Conclusions ... 96
3.2 DIFFERENTIAL PERSISTENCE PATTERNS OF OPIOID USE AMONG PATIENTS WITH NON-CANCER AND CANCER PAIN OVER A THRee- YEAR FOLLOW-UP PERIOD ... 97
3.2.1 Introduction ... 97
3.2.2 Aims and Objectives ... 99
3.2.3 Methods ... 100
3.2.4 Results ... 105
3.2.5 Discussion ... 114
3.2.6 Strengths and Limitations ... 120
3.2.7 Conlcusion ... 121
CHAPTER FOUR: CLINICAL OUTCOMES AND RISKS OF OPIOID ABUSE/MISUSE AMONG SHORT-TERM AND LONG-TERM OPIOID USERS AMONG PATIENTS WITH NON-CANCER PAIN ... 122
xi
4.1 CLINICAL OUTCOMES OF SHORT-TERM AND LONG-TERM OPIOID
USE AMONG PATIENTS WITH NON-CANCER PAIN ... 122
4.1.1 Introduction ... 122
4.1.2 Aims And Objectives ... 123
4.1.3 Methods ... 124
4.1.4 Results ... 156
4.1.5 Discussion ... 189
4.1.6 Strengths and Limitations ... 202
4.1.7 Conclusion ... 202
4.2 FACTORS ASSOCIATED WITH HIGH-RISK OF OPIOID ABUSE OR MISUSE AMONG PATIENTS WITH NON-CANCER PAIN ... 204
4.2.1 Introduction ... 204
4.2.2 Aims and objectives ... 207
4.2.3 Methods ... 208
4.2.4 Results ... 215
4.2.5 Discussion ... 237
4.2.6 Strengths and limitations ... 245
4.2.7 Conclusion ... 246
CHAPTER FIVE: GENERAL DISCUSSION AND CONCLUSIONS ... 247
5.1 Implications for policy, practice, and future research ... 250
REFERENCES ... 252
APPENDIX I: ETHICAL APPROVAL ... 287
APPENDIX II-A: CANCER PAIN PATIENTS’ DIAGNOSES GROUPING ... 290
APPENDIX II-B: NON-CANCER PAIN PATIENTS’ DIAGNOSES GROUPING ... 306
APPENDIX III: MISCELLANEOUS OPIOID PRESCRIBER DISCIPLINES ... 313
APPENDIX IV-A: PATIENT INFORMATION SHEET (ENGLISH) ... 3184
APPENDIX IV-B: PATIENT INFORMATION SHEET (BAHASA MALAYSIA) ... 32416
APPENDIX V-A: CONSENT FORM (ENGLISH) ... 318
APPENDIX V-B: CONSENT FORM (BAHASA MALAYSIA) ... 321
APPENDIX VI: CASE REPORT FORM (CRF) ... 324
APPENDIX VI-A: BRIEF PAIN INVENTORY-SHORT FORM (BPI-SF) ... 331
APPENDIX VI-B: SCREENER AND OPIOID ASSESSMENT FOR PATIENTS WITH PAIN-REVISED ... 332
APPENDIX VI-C: SHORT FORM-36 VERSION 2 HEALTH SURVEY ... 334
APPENDIX VI-D: OPIOID SIDE EFFECTS ASSESSMENT ... 340
APPENDIX VI-E: ASSESSMENT OF ADVERSE EFFECTS ASSOCIATED WITH LONG-TERM OPIOID USE ... 341
APPENDIX VII-A: BPI-SF LICENSE AGREEMENT ... 342
APPENDIX VII-B: SOAPP-R LICENSE AGREEMENT... 3424
APPENDIX VII-C: SF-36V2 LICENSE AGREEMENT ... 3428
xii
APPENDIX VIII: MEDICATION CLASS AND DRUGS INCLUDED FOR MQS ANALYSIS ... 355 APPENDIX IX: HEALTH-RELATED QUALITY OF LIFE OF THE GENERAL MALAYSIAN POPULATION ... 356 APPENDIX X: RESULTS OF ANALYSIS OF ALL SOAPP-R ITEMS ... 357 APPENDIX XI: NORMALITY OF DATA FOR DAILY OPIOID DOSE ... 34258 APPENDIX XII: NORMALITY OF DATA FOR DAYS COVERED WITH OPIOIDS ... 342
xiii
LIST OF TABLES
Table 2.1 Examples of instruments used to assess the risk of opioid abuse or misuse 46 Table 3.1 Equianalgesic ratio for opioids included in this study. 59
Table 3.2 Percentage of missing data and data imputed. 64
Table 3.3 Patient demographics 71
Table 3.4 Means and medians of daily opioid dose (mg morphine equivalents) per patient over the quarters of 3 follow-up years in cancer and non-cancer groups. 79 Table 3.5 Mean (SD) and median (IQR) of total days covered with opioids per patient
throughout study duration in cancer and non-cancer groups. 80 Table 3.6 Means and standard deviations (SD) of total days covered with opioids per
patient over the quarters of 3 follow-up years in cancer and non-cancer groups.
82 Table 3.7 Definitions of persistent opioid use. Cut-off points per 365 days and typical
clinical scenarios connected with each definition of persistent opioid use. 101
Table 3.8 DDD values for opioids included in the study. 102
Table 3.9 Demographics of non-persistent and persistent opioid users among non-cancer
and cancer patients. 107
Table 4.1 Summary of the eight health domain scales of the SF-36v2 instrument. 137
Table 4.2 Charlson Comorbidity Index 141
Table 4.3 Detriment Weights of Pain Medication Classes 144
Table 4.4 Relative Dosage Level Scores 145
Table 4.5 Dependent and independent variables included in logistic regression analysis in
this study. 151
Table 4.6 Patient Demographics 158
Table 4.7 Percent frequency of the type of opioids prescribed to patients in this study. 160 Table 4.8 Percent frequency of the proportions of opioid prescriptions for short-term and
long-term opioid users. 161
xiv
Table 4.9 Percent frequency and chi-square test of the type of opioid prescribed to short-
term and long-term users. 163
Table 4.10 Median days supplied with opioids throughout the study period among short-
term and long-term opioid users. 164
Table 4.11 Comparison of the total opioid dose prescribed overall between short-term and
long-term opioid users. 165
Table 4.12 Comparison of the opioid dose prescribed per day per patient throughout the study period between short-term and long-term opioid users. 166 Table 4.13 Total number of medications prescribed to short-term and long-term users
according to medication classes during the study period. 169 Table 4.14 Comparison of Total MQS-III score per patient between short-term versus
long-term opioid users at index prescription date (t=0) and subsequent yearly time points (t=1, t=2, t=3, t=4) during the study period. 171 Table 4.15 Total number of medications prescribed in each class at time of clinical
outcomes assessment among short-term and long-term opioid users. 173 Table 4.16 Comparison of ‘current’ Total MQS scores per patient between short-term and
long-term users. 174
Table 4.17 Total number of patients and the relative detriment of the medication classes
prescribed among short-term opioid users (n=30). 175
Table 4.18 Total number of patients and the relative detriment of the medication classes
prescribed among long-term opioid users. 176
Table 4.19 Means and standard deviations of Brief Pain Inventory Items between short-
term and long-term opioid users. 178
Table 4.20 Scores of SF-36v2 health domain scales and summary measures between
short-term and long-term opioid users. 180
Table 4.21 Comparison of the proportion of patients reporting each side effect between short-term (n = 30) and long-term users (n = 31). 182 Table 4.22 Side Effects Rating Scores among those reporting the relevant side effects
between short-term and long-term opioid users. 183
Table 4.23 Proportion of short-term (n = 30) and long-term (n = 31) opioid users who reported adverse effects associated with long-term opioid use in this study. 185 Table 4.24 Univariable and multivariable binary logistic regression of variables
associated with long-term opioid use. 187
xv
Table 4.25 Variables selected via forward selection approach (stepwise logistic
regression) in the final model. 189
Table 4.26 Dependent and independent variables included in logistic regression analysis
in this study. 213
Table 4.27 Patient demographics 217
Table 4.28 Differences between Low-risk and High-risk patients on individual items of
the SOAPP-R. 219
Table 4.29 Comparison of the SOAPP-R domains between low-risk and high-risk patients 220 Table 4.30 Comparison of the proportion of each type of opioid (%) prescribed between
low-risk and high-risk patients. 221
Table 4.31 Number of patients and Mean days covered with opioids per patient among
low-risk and high-risk patients 223
Table 4.32 Number of patients and Mean opioid daily dose per patient (OMEQ)
prescribed in each follow-up year among low-risk and high-risk patients 225 Table 4.33 Proportion of patients by mean opioid daily dose ranks (OMEQ) in each
follow-up year among low-risk and high-risk patients. 227 Table 4.34 Means and standard deviations of Brief Pain Inventory (BPI) Items between
low-risk and high-risk patients. 229
Table 4.35 Scores of SF-36v2 health domain scales and component summary measures
between low-risk and high-risk patients. 232
Table 4.36 Univariable and multivariable binary logistic regression of variables
associated with high risk of opioid abuse or misuse. 235
xvi
LIST OF FIGURES
Figure 1.1 Loeser's multifaceted model of the components of pain 7
Figure 1.2 WHO three-step analgesic ladder 15
Figure 1.3 Thesis structure 31
Figure 2.1 An 11-point Numerical Rating Scale (NRS) 40
Figure 3.1 Example for calculation of total days covered with opioids with overlapping days between two prescriptions of a patient in a follow-up period. 61
Figure 3.2 Data preparation of study dataset 63
Figure 3.3 Common diagnoses of patients prescribed opioids in cancer group (n=665). 72 Figure 3.4 Common diagnoses of patients receiving opioids in non-cancer group (n=256)
73 Figure 3.5 Total number of opioid prescriptions (n = 3,323) prescribed during 2013-2015.
74 Figure 3.6 Number of prescriptions (n = 3,323) for cancer and non-cancer pain during
2013-2015. 74
Figure 3.7 Types of opioid prescriptions prescribed for cancer and non-cancer groups. 75 Figure 3.8 Disciplines issuing opioid prescriptions in cancer and con-cancer groups. 77 Figure 3.9 Mean daily opioid dose (mg morphine equivalents) per patient over the
quarters of 3 follow-up years in cancer and non-cancer groups. 78 Figure 3.10 Mean total days covered with opioids per patient over the quarters of 3
follow-up years in cancer and non-cancer groups. 81
Figure 3.11 Change in persistence pattern among persistent opioid users with non-cancer
pain over three follow-up years. 111
Figure 3.12 Change in persistence pattern among persistent opioid users with cancer pain
over three follow-up years. 113
Figure 4.1 Outline of the study design. 124
Figure 4.2 Flow chart of the process of patient recruitment 126
xvii
Figure 4.3 Example of short-term (<90 days) and long-term (≥90 days) opioid user. 129
Figure 4.4 BPI Measurement Model 134
Figure 4.5 SF-36v2 Health Survey Measurement Model. 138
Figure 4.6 SF-36v2 scoring software (Home page) 139
Figure 4.7 SF-36v2 scoring software (Project page) 140
Figure 4.8 Charlson Comorbidity Index Calculator 142
Figure 4.9 Sample of MQS III calculation for a particular patient. 145 Figure 4.10 Type of opioid prescriptions (n=653) prescribed to recruited patients during
the study period. 161
Figure 4.11 Proportion of the total opioid prescriptions (n = 653) prescribed to short-term
and long-term opioid users. 162
Figure 4.12 Comparison of the proportion of each type of opioid (%) prescribed between
short-term and long-term opioid users. 163
Figure 4.13 Comparison of the proportions of patients prescribed each daily opioid dose rank between short-term and long-term opioid users. 167 Figure 4.14 Comparison of proportions of medication classes between short-term versus
long-term opioid users. 170
Figure 4.15 Boxplot of Total MQS-III scores at index prescription date (t=0) and at subsequent yearly time points among short-term vs. long-term users during
study period. 172
Figure 4.16 Comparison of SF-36v2 scores of short-term and long-term opioid users with the Malaysian adult population aged ≥18 years old. 181 Figure 4.17 The proportion of each type of opioid (%) prescribed among low-risk and
high-risk patients. 221
Figure 4.18 Mean days covered with opioids per patient in each follow-up year among
low risk and high risk patients 222
Figure 4.19 Mean opioid daily dose per patient (OMEQ) prescribed in each follow-up
year among low-risk and high-risk patients 224
Figure 4.20 Proportion of patients by mean opioid daily dose rank (OMEQ) in each follow-up year among low-risk and high-risk patients 226
xviii
Figure 4.21 Comparison of SF-36v2 scores of low-risk and high-risk patients with the Malaysian adult population aged 18 years and above. 233
xix
LIST OF ABBREVIATIONS
95% CI 95% Confidence Interval
ADRBs Aberrant drug-related behaviours BPI-SF Brief Pain Inventory – Short Form CCI Charlson Co-morbidity Index
CDC US Centers for Disease Control and Prevention CNCP Chronic Non-Cancer Pain
DDD Defined Daily Dose
DIRE Diagnosis, Intractability, Risk, and Efficacy Score GOF Goodness-Of-Fit
HKL Hospital Kuala Lumpur
HRQoL Health-Related Quality of Life HTAA Hospital Tengku Ampuan Afzan
IASP International Association for the Study of Pain ID Identification
IQR Inter-Quartile Range LR Logistic Regression
LTOT Long-Term Opioid Therapy MCS Mental Component Summary MME Morphine Milligram Equivalents MQS-III Medication Quantification Scale - III NRIC National Registration Identity Card NSAIDs Non-Steroidal Anti-Inflammatory Drugs OMEQ Oral Morphine Equivalent
OR Odds Ratio
ORT Opioid Risk Tool
PCS Physical Component Summary PRN Pro Re Nata / "As Needed"
QoL Quality of Life
RCT Randomized Controlled Trials
S-DDD Defined Daily Doses for statistical purposes SD Standard Deviation
SF-36v2 Short Form - 36 version 2 Health Survey
SOAPP-R Screener and Opioid Assessment for Patients with Pain – Revised TCA Tricyclic Anti-Depressants
WHO World Health Organization
xx
CONFERENCE PRESENTATIONS
Asween Rowena Abdullah Sani, Che Suraya Zin, Zaswiza Mohd Noor, Abdul Hadi Mohamed, Lisa Nissen. Differential patterns of persistent opioid use in patients with cancer and non-cancer pain. Value in Health, 2016;19 (Issue 7): A814. Oral presentation, The International Society for Pharmacoeconomics and Outcomes Research (ISPOR) 7th Asia- Pacific Conference, 4th-6th September 2016 at Suntec Convention Centre, Singapore.
Asween Rowena Abdullah Sani, Che Suraya Zin, Abdul Hadi Mohamed, Munira Izat, Tan Hung Ling, Ng Kim Swan. Persistence patterns of opioid use over 3 follow-up years: The difference between non-cancer and cancer pain. Poster presentation, The 6th Biennial Scientific Meeting 2018 of the Malaysian Association for the Study of Pain, 17th – 18th March 2018 at National Cancer Institute, Putrajaya, Malaysia.
LIST OF PUBLICATIONS
Sani, Asween R., Zin, Che Suraya, Mohamed, Abdul H., Izat, M., Tan, Hung L., Ng, Kim S., & Nissen, L. (2019). Exploration of change in persistence patterns of opioid use among patients with non-cancer and cancer pain over a three-year follow-up period. Journal of Pharmacy Practice and Research, 50(1). pp. 28-35. doi:10.1002/jppr.1573
1
CHAPTER ONE 1 INTRODUCTION
1.1 BACKGROUND OF THE STUDY
Studies showing increasing trends of opioids prescribed for non-cancer pain especially chronic non-cancer pain (CNCP) are extensive in literature despite the scarcity of evidence supporting the effectiveness of opioid use with regards to improved analgesia and functionality in these patients (Bosetti et al., 2019; Chou et al., 2015; Curtis et al., 2019;
Hollingworth et al., 2015; Mojtabai, 2018). At the same time, increasing evidence of opioid-related harms such as endocrine deficiencies, cardiovascular events, abuse, addiction, overdose, and overdose-related deaths began to emerge in non-cancer patients particularly those on long-term opioid therapy (Baldini et al., 2012; Chou et al., 2015).
Nonetheless, there is a legitimate need for opioids in especially CNCP patients and literature suggests long-term opioid therapy may benefit a subgroup of carefully selected and monitored CNCP patients (Noble et al., 2010; Saïdi et al., 2018). Consequently, opioid prescribing guidelines have been developed and improvised over the years which adopt an individualized approach to ensure appropriate opioid prescribing and proper selection of patients where the benefits of opioid therapy outweigh risks (Ballantyne, 2015a; Jason W.
Busse et al., 2017; Chou, Fanciullo, Fine, Adler, et al., 2009; Dowell et al., 2016).
However, most of these researches on opioid use patterns and risks were conducted in developed countries such as the US, Canada, and Nordic countries, and opioid prescribing guidelines were developed based on evidences from these studies. Little is
2
known on opioid prescribing patterns in Malaysia as there is limited research in the patterns of opioid use particularly among the non-cancer pain population in this country. The available aggregate data on national opioid consumption in this country is inadequate to provide information on actual patterns of opioid use in clinical practice and at the patient- level. Moreover, a majority of the research on opioid use in this country is focused on illicit drug abusers under methadone maintenance therapy. There is also a lack of information on the characteristics of non-cancer patients prescribed opioids for long-term and on the patient characteristics vulnerable to opioid-related harms such as misuse, abuse, and addiction in this region.
In this light, this thesis was conducted to add to the body of knowledge of appropriate opioid prescribing specifically for the better management of non-cancer pain involving opioids. This thesis investigated patterns of opioid use including the differential patterns of persistent opioid therapy at the patient level in cancer and non-cancer pain population at outpatient tertiary hospital settings. This thesis further focused on patients with non-cancer pain at pain clinic settings by evaluating the clinical outcomes of short- term and long-term opioid use and also identified risk factors associated with low risk and high risk of opioid abuse or misuse in these patients.
3 1.2 PAIN
The International Association for the Study of Pain (IASP) defines pain as “an unpleasant sensory and emotional experience associated with actual or potential tissue damage, or described in terms of such damage” (IASP Terminology Working Group, 2017). Pain is inherently subjective, a multidimensional experience as a result of various influences ranging from genetic predispositions to environmental, social, psychological, and sensory factors (Melzack & Katz, 2013). The sensation of pain is not simply an end product of activation of pain receptors as a result of injury, inflammation, or other tissue pathology but it is a complex process involving generation of neural signals influenced by past experience, culture, environmental and personal factors (Melzack & Katz, 2013).
Essentially, pain serves as a warning signal for survival but can also be a disease in itself.
1.2.1 Pathophysiology of Pain
Pain can be classified based on its underlying pathophysiology as nociceptive and/or neuropathic pain. Nociceptive pain is pain generated from the activation of nociceptors at the periphery due to injury or tissue damage (Basbaum et al., 2009; Gangadharan & Kuner, 2013; R. D. Treede, 2018). Nociceptors are free nerve endings of primary afferent sensory neurons found in the peripheral nervous system which convert noxious stimuli such as tissue damage into electrical impulses (Ellison, 2017). These electrical impulses are then transmitted to the central nervous system which the brain interprets to produce pain sensations (Ellison, 2017). Nociceptive pain is further subdivided into somatic or visceral pain (Basbaum et al., 2009). Somatic pain is pain arising from tissues such as skin, muscle, joints, and bones. It is described as aching, stabbing, gnawing, or throbbing and can either
4
be intermittent or constant (Anwar, 2016). Visceral pain is pain arising from visceral organs which are not sensitive to pain (Anwar, 2016; Ellison, 2017). It is also known as
“referred pain” because pain is usually perceived as occurring in a region of the body which is either remote or adjacent from the actual source of pain1. Visceral pain is often described as dull, aching, and diffuse (Anwar, 2016; Visser & Davies, 2009).
Neuropathic pain is pain arising from lesion or disease affecting the sensory nervous system (Ellison, 2017). Neuropathic pain is commonly caused by metabolic disorders (e.g.
painful diabetic neuropathy), infection (e.g. HIV), nerve compression, inflammation, trauma, and tumors (Colloca et al., 2017). It is usually described as sharp and burning (Colloca et al., 2017). Neuropathic pain is generally classified as peripheral (damage to peripheral nerve, plexus, dorsal root ganglion, or root) or central (damage to the brain or spinal cord) (Colloca et al., 2017; Ellison, 2017).
1.2.2 Acute pain versus Chronic pain
Pain can also be classified based on its time course and duration as either acute or chronic pain. Acute pain has been defined as “pain of recent onset and probable limited duration;
it usually has an identifiable temporal and causal relationship to injury or disease” (Ready et al., 1992). It is self-limiting and typically lasts less than 3 months (Anwar, 2016; Ellison, 2017). It is usually nociceptive and resolves upon healing of the underlying tissue injury (Anwar, 2016). Acute pain is commonly due to surgery, traumatic injury, tissue damage, medical procedures, and acute disease states (McCormick & Law, 2016). Acute pain is vital for survival as it provides warning signals of potential injury and the magnitude of an injury.
This biological significance is evident in individuals with “congenital insensitivity to pain”,