Examining distinct patterns of time-use and their associations with quality of life in individuals receiving treatment for opioid use disorder

Thank you to the authors for revising their registered report so thoroughly. The revisions have taken account of all my concerns with the original draft and I now find this report a robust planned study. I look forward to reading the results in due course. 

The authors have responded thoroughly and satisfactorily to all my comments. I wish them the best of luck as they progress with the study.

1A. The scientific validity of the research question(s): 

The research questions are scientifically valid and are exploratory in nature. I woudl have also liked to see research questions handling the specific anticipated findings for each medication for opioid use disorder. I found the concept of "time use" to be a little woolly and thought that the authors could better articulate this and group in to substance related versus recovery related. I also thought that rather than just looking at the quantity of the activity (how many days followed up by a rating of the quality of that aspect). 

1B. The logic, rationale, and plausibility of the proposed hypotheses (where a submission proposes hypotheses): 

While the hypotheses are logical to someone with knowledge in the area, I do not think that at present they are derived from what precedes them. You might imagine that type of medication would be related to time use pattern because time spent travelling to daily supervised consumption vs. a single long acting injection would dictate this. If the authors are implying that other factors affect this, they need to clearly provide a rationale in the introduction. Similarly with stigma, mental health and quality of life, a better case needs to be made for this relationship. 

1C. The soundness and feasibility of the methodology and analysis pipeline (including statistical power analysis or alternative sampling plans where applicable): 

The methodology is sound and feasible, but I think it excludes a large proportion of people with OUD because it a) required people to be able to read and consent and b) required digital completion of questionnaires via a smartphone or device. Many people in recovery from OUD may not have access to digital tech, and thus the sample recruited and retained will be skewed heavily towards more stable people who might naturally make better use of their time and have higher wellbeing, and experience lower stigma. 

1D. Whether the clarity and degree of methodological detail is sufficient to closely replicate the proposed study procedures and analysis pipeline and to prevent undisclosed flexibility in the procedures and analyses: 

Yes, this is detailed and would be easy to replicate. 

1E. Whether the authors have considered sufficient outcome-neutral conditions (e.g. absence of floor or ceiling effects; positive controls; other quality checks) for ensuring that the obtained results are able to test the stated hypotheses or answer the stated research question(s).

See above regarding exclusion of groups without access to digital tech. The statistical methods are robust, but modifying the sampling and data collection protocol, and clearly stating how group (i.e. type of medication) will be handled in the analysis would improve this. 

Other points: 

Abstract: 

"Frees up considerable time" is a very vague statement when referring to benefits of long acting Buprenorphine. For who? In what way? Also amount of time does not necessarily = increased quality of time. 

This is an interesting proposal to explore the potential effects of long-acting medications in opioid use disorder, focusing in particular on the time saving experienced by patients and any consequent association with outcome measures including well-being, experienced stigma, and life satisfaction. I am not an expert in this topic and am providing a non-specialist review to evaluate the proposal in accordance with more general RR requirements. Overall, though, I thought it tackled a valid and interesting question using a useful approach (e.g. the use latent profile analysis is unusual for RRs).

I offer some comments below to help the authors refine and optimise their current plan.

1. Perhaps my most substantive concern is that the design appears to lack a statistical sampling plan. The authors plan to recruit 500 patients (the minimum apparently required for the latent profile analysis), and their sampling plan is understandably limited by available resources; however, given the intended use of null hypothesis significance testing to evaluate hypotheses H2-H5, the design lacks the necessary sensitivity power analysis to confirm that this sample size is adequate. The authors really need to show what effect size this maximally obtainable sample size has sufficient power to detect for each of these predictions, and to then justify this minimally detectable effect accordingly (on a related note, the entries in the design table column “Rationale for deciding the sensitivity of the test for confirming or disconfirming the hypothesis” do not answer the question asked by this section of the table).
2. The hypotheses are non-directional, which suggests that there might be reasons to expect a negative association between long-acting medication use and the various outcome measures. Is this the case, and if so, could the authors explore the rationale for predicting positive or negative associations in the introduction? With my non-specialist hat on, it seemed intuitive for the hypotheses to be directional (predicting benefits), but perhaps there are reasons to expect effects in either direction.
3. Data will be collected by digital format only. How severe a sampling bias does this introduce for this particular patient population? I think some additional discussion of this concern is warranted in the main text, ideally with reference to previous literature, e.g. considering the rate with which patients in this group have access to the necessary technical infrastructure to participate. I would also recommend including a Constraints on Generality section to ensure that all conclusions are appropriately calibrated in advance (see https://journals.sagepub.com/doi/10.1177/1745691617708630)
4. Please provide a link to the eSupplement in the main text (https://osf.io/5evuh)
5. The Stage 1 manuscript is written in past tense but is also entered as Level 6 in the submission checklist. I assume this is just a stylistic convenience to minimise text changes at Stage 2 (which is fine) but please confirm for sure in the response to reviewers whether any of the data has (yet) been collected. If some data has already been collected, this must be stated in the Stage 1 manuscript and the bias control level adjusted down accordingly. If no data has been collected (and none will be until after IPA, which I assume is the case) then a comment to this effect in the response to reviewers will suffice.