PATTERNS OF OPIOID USE IN PATIENTS WITH CANCER AND NON-CANCER PAIN AND ITS

(1)

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

(2)

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.

(3)

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 .

(4)

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

(5)

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 ………..

(6)

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.

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

(7)

vii

This thesis is dedicated to my beloved parents, Abdullah Sani and Adeda Rahimah, for their endless love, unconditional support, and encouragement.

(8)

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.

(9)

ix

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

(10)

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

3.2.2 Aims and Objectives ... 99

3.2.3 Methods ... 100

3.2.4 Results ... 105

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

(11)

xi

4.1 CLINICAL OUTCOMES OF SHORT-TERM AND LONG-TERM OPIOID

USE AMONG PATIENTS WITH NON-CANCER PAIN ... 122

4.1.2 Aims And Objectives ... 123

4.1.3 Methods ... 124

4.1.4 Results ... 156

4.1.7 Conclusion ... 202

4.2 FACTORS ASSOCIATED WITH HIGH-RISK OF OPIOID ABUSE OR MISUSE AMONG PATIENTS WITH NON-CANCER PAIN ... 204

4.2.2 Aims and objectives ... 207

4.2.3 Methods ... 208

4.2.4 Results ... 215

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

(12)

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

(13)

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

(14)

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

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

(15)

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

(16)

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

(17)

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

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

(18)

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

(19)

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

(20)

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

(21)

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

(22)

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.

(23)

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

(24)

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”,

PATTERNS OF OPIOID USE IN PATIENTS WITH CANCER AND NON-CANCER PAIN AND ITS