Table 2 |.
Items to report and recommendations in the General fMRI Information category (category 2) of the checklist
| Subcategories | Main items to report | Item importance (1–5) | Specific recommendation to report | Recommendation inclusion |
|---|---|---|---|---|
|
| ||||
| fMRI Pulse Sequence and Other Acquisition Details | 2.1. fMRI data-acquisition details | 4.77 (0.52) | 2.1.1. Report fMRI data acquisition details based on the available checklists (e.g., COBIDAS). FMRI data acquisition details might have explicit effects on drug-cue reactivity results, for example, number of head coil channels, because higher channels (32 compared to 8) might be associated with better SNR in cortex, with the cost of losing signal in the deep parts of the brain | 47 (98%) |
| fMRI Preprocessing Pipeline and Other Details | 2.2. fMRI preprocessing details | 4.81 (0.45) | 2.2.1. Report fMRI preprocessing details based on the available checklists (e.g., COBIDAS). There are items in the preprocessing steps that might have an effect on fMRI drug-cue reactivity results. For example, higher FWHM might be related to the loss of signal in small nuclei 2.2.2 Report motion differences between participant groups (i.e., individuals with an SUD vs. controls), because higher motion during the drug-related blocks compared to neutral blocks might act as a confounder 2.2.3 Report quality-control measures, artefact detection methods and the threshold to exclude participants with heavy movement |
48 (100%) 38 (79%) 47 (98%) |
| fMRI Data Processing | 2.3. fMRI analyses and statistical modeling details | 4.89 (0.31) | 2.3.1. Report fMRI single-subject-level and group- level processing steps on the basis of the standard checklists (e.g., COBIDAS) 2.3.2. Report whether GLM analyses are random, mixed,or fixed effects for inclusion in future metaanalyses 2.3.3 Report all covariates used for each model and whether or not demeaning was done for covariates of interest 2.3.4 Report any publicly available tool/software use (e.g., SPM, AFNI and FSL) 2.3.5 Report any attempt for preregistration of data processing methods 2.3.6 Report methods that are used to control for multiple comparisons error and spatial autocorrelations 2.3.7 Report the definition of the ROIs for studies using an ROI approach 2.3.8 Provide effect sizes for all reported statistics |
47 (98%) 47 (98%) 47 (98%) 47 (98%) 39 (83%) 47 (100%) 47 (100%) 33 (73%) |
| fMRI Data Reporting | 2.4. Basic whole-brain response to drug cues | 4.38 (0.68) | 2.4.1 Report the second-level maps or activation foci therein of each study group singly, as well as a group-difference map (e.g., between the clinical group and the control group) (if applicable) in the results or the supplements as a figure or table (foci coordinates and stats) with details on the thresholding measures and quantities. Even if the paper has other analyses (e.g., task-based connectivity), the whole-brain maps of the craving>neutral contrast should be reported for comparison with other studies and future metaanalyses 2.4.2 Report beta values for both conditions (craving and neutral), because an ‘activation’ in the mPFC during craving could be explained by a de-activation in the control condition 2.4.3 Report the contrast map for other included conditions (e.g., multiple drug stimuli, affective images and other active control) if other conditions are included 2.4.4. Provide the effect size map, the non-thresholded statistical map and the data in an accessible repository (e.g., OSF, NIMH/NIAAA data archive, GitHub orNeurovault) 2.4.5. It is understandable that researchers who are not using conventional whole-brain GLM-based methods (e.g., ICA, Graph Theory, PPI connectivity, ROI only analysis) or developing other innovative and non-conventional methods might face difficulties in reporting ‘whole-brain response to drug cues’. It is still recommended for these studies to consider strategies for reporting whole-brain responses to drug cues to make data/results aggregation and comparison possible |
45 (96%) 41 (85%) 31 (66%) 32 (70%) 37 (82%) |
| General Recommendations | - | - | 2.0.1. Refer to standard checklists (e.g., COBIDAS) for items in this category. Items in the ENIGMA ACRI checklist are designed to be dichotomous (Yes or No); however, there is a continuum for the details to be reported. Provide as much detail as available | 28 (78%) |
Ratings for items (1–5) are reported as mean (s.d.) in the ‘Item importance’ column, and ratings for recommendations (Yes) are reported as frequency (percentage%) of ‘Yes’ reports in the ‘Recommendation inclusion’ column. AFNI, Analysis of Functional Neuroimages; FSL, FMRIB Software Library; GLM, generalized linear model; ICA, independent component analysis; mPFC, medial prefrontal cortex; NIAAA, National Institute on Alcohol Abuse and Alcoholism; NIMH, National Institute of Mental Health; OSF, Open Science Foundation; PPI, psychophysiological interaction; ROI, region of interest; SNR, signal-to-noise ratio; SPM, statistical parametric mapping.