Translation and Validation of the COMM and ASI-SR

There is no validated instrument for the assessment of early detection of substance use disorder development in swedish language.

In order to address this gap, investigators plan to translate and validate the COMM questionnaire using the ASI-SR in a Swedish population of pain patients undergoing opioid treatment. In addition, study will investigate the prevalence of illicit substance use in this population and compare it to patients who are not receiving opioid treatment. The collection of material (blood) in the Uppsala Biobank for future analysis of neurobiological processes during long-term pain treatment will be performed.

Phase 1:

Translation of the instrument takes place according to the "Patient-Reported Outcome (PRO) Consortium translation process guidelines"and according to guidelines from the company MAPI ResearchTrust, which owns the rights to use of the instrument.

Phase 2: The subsequent phase entails validating the COMM instrument against the ASI-SR in a Swedish sample of patients experiencing long-term pain, in accordance with the "Cosmin guidelines" [40], and involves the following procedures. Firstly, the reliability of the COMM instrument by examining internal consistency, test-retest consistency, standard error of measurement (SEM), and conducting factor analysis with rotation in a Swedish population of patients with long-term pain andlong-term opioid treatment (LOT)will be assessed. Secondly, validity of the COMM instrument with regard to comprehensiveness, scope, user-friendliness, and predictability within this particular population will be investigated. Furthermore, investigators will assess the acceptability of the instrument among the study population. Finally, exploration of the prevalence of substance use alongside LOT in pain patients and compare it with pain patients who do not receive opioid treatment will be performed. Inclusion goal is to recruit 200 research subjects from various healthcare settings, including primary care, specialized pain care, other specialized care (e.g., Endometriosis Center), and addiction care (addiction medicine program for pain patients) for the test group. Additionally, 200 individuals are planned to be recruited from pain centers and primary care for the control group to measure prevalence rates. The control group will consist of patients with long-term pain who are not undergoing opioid treatment. Since the study involves two groups of patients with long-term pain (with and without opioid treatment), research group will inquire separately if participants are willing to provide blood samples for biomarker analysis of pain (O-link panel) and genetic analysis, which will be stored in the biobank for future research on the effects of opioids.

Validation will be conducted using the results of drug and alcohol analysis from salivatests and blood. Each category, as identified by the original COMMinstrument, will be evaluated against the following instruments:

1. Symptoms and signs of drug abuse (question 1) - AUDIT, DUDIT, ASI-SR, drug and alcohol test
2. Emotional and psychiatric problems (questions 2, 5, 7, 8, 13) - ASI-SR, GAD-7, PHQ9
3. Visitor characteristics/patterns (questions 3, 12, 17) - medical record data
4. Occurrence of untrue statements (doubt/lie/untruth) and drug use (questions 4, 6, 9, 10, 11) - medical record data, drug and alcohol tests
5. Medication use and deviation from prescription (non-compliance) (questions 14, 15, 16) - medical record data, occurrence of overprescription. Pain-related questions will be assessed using the BPI-SF.

In addition, research subjects will also be asked if they would like to be interviewed by a researcher in connection with their consent. Thirty of the initial research subjects who respond positively will undergo a semi-structured interview, where ease of use will be examined using the "think aloud" method. This involves asking subjects about their thoughts on the forms, including their understanding of the questions, language choices, and any possible negative associations. The interviews will then be reviewed using qualitative content analysis.

Moreover, an extensive assessment will be undertaken, involving the input of 10 esteemed professionals in Sweden who specialize in various fields related to this subject matter, such as pain doctors, addiction medicine doctors, psychologists, behavioral scientists, and others. These experts will be approached to provide their valuable insights by responding to a set of carefully devised questions concerning the evaluation of content validity. Their estimations will serve as the fundamental basis for calculating the Content Validity Index (CVI).

In order to collect and analyze the health data, information from medical records, relevant forms, and laboratory responses are compiled into a study database and then transferred to a statistics program. To ensure internal consistency and reliability, Cronbach's alpha is used, with the expectation that it will be higher than 0.70 at a significance level of p<0.05. Additionally, the agreement between the two test occasions of COMM is assessed using the intraclass correlation coefficient (ICC). The ICC is classified as satisfactory if below 0.40, good if between 0.40-0.59, and excellent if between 0.60-0.70. The correlation between the results of COMM and the ASI-SR is examined using both the Pearson and Spearman correlation coefficients, which are calculated for the respective clusters of questions in the form. Furthermore, confirmatory factor analysis with rotation is employed to investigate the relationship between different clusters or dimensions of questions that can be identified for correlation purposes. The discriminant validity of ASI-SR composite scores and external validity measures are determined by calculating the intercorrelation matrix between COMM groups of questions, ASI-SR composite scores, and external validation measures. The minimum requirement for correlation is set at >0.40. Statistical processing is conducted using SPSS, SAP, and Amos programs. It should be noted that the qualitative aspect of the study is not covered by the statistical plan. In terms of prevalence estimation, both the test group and the control group will be described using descriptive statistics. The occurrence of addictive drugs, non-prescribed drugs, and measures of alcohol use (PETh) during the last three weeks in both groups will be compared using a chi-square test. Odds analysis will be performed to assess secondary use of illegal substances, parallel dependence on other substances, and drugs and alcohol during the length of the study.