A prescription exists, but the next pharmacy cannot dispense it. A referral is recorded, but nobody has accepted the person into care. A service remains open, but its new hours make it unreachable.
These examples reveal the question behind Safety Continuity: does the protection people need remain usable while the risk continues?
The Real DOPE Project proposes Safety Continuity as a way to examine protection across time and transitions. Its starting point is practical: the existence of a service and a person's ability to use its protective function are different things. A change in provider can preserve protection when a suitable replacement works in time. Keeping the same organization can still leave a gap.
This is a conceptual research proposal. The framework has not yet been validated, and this article reports no new participant data.
What Safety Continuity means
TRDP's current research protocol proposes this definition:
Safety Continuity is the degree to which a person or population retains timely, usable access to risk-appropriate protective capability while a material risk persists, including across changes in setting, service, institutional or legal status, policy, or hazard conditions.
In everyday terms, identify a risk, identify a protection that has evidence relevant to that risk, and follow what happens to that protection over time.
Four features help make the question concrete:
- Availability: Is the resource operating when it is needed?
- Usability: Can this person reach and use it, considering cost, transport, hours, accessibility, eligibility and privacy?
- Adequacy: Does it meet the particular need?
- Connection: Does the next necessary step actually happen?
A protective function must be specified before examining the outcome. Survival does not prove that protection was adequate. Harm does not, by itself, prove that continuity failed.
What the idea builds on
Safety Continuity draws on established scholarship. Haggerty and colleagues describe informational, management and relational continuity of care. Levesque and colleagues describe access through the interaction of health services and people's ability to recognize a need, seek help, reach it, pay for it and engage with appropriate care. [1][2]
TRDP does not claim to have invented continuity or discovered that access matters. Its proposed contribution is to bring interruption timing, duration, adequacy, replacement and restoration into one explicit examination of protective functions across institutional and community boundaries.
That contribution needs testing. If existing access or continuity measures answer the question just as well with less burden, the new framework may be unnecessary. TRDP's review protocol allows for that result: Safety Continuity might prove to be a distinct construct, a useful synthesis, an implementation framework, or a concept that should be retired.
How protection becomes disconnected
Within the wider TRDP research, Selective Emergency Theory asks whether unequal institutional recognition and prioritization help explain differences in protection. Safety Continuity follows whether protection persists. Safety Distribution organizes questions about where and how protection reaches people.
The five proposed Safety Distribution domains make those questions easier to investigate:
| Domain | Question to examine |
|---|---|
| Access | Can the person actually receive the relevant protection? |
| Timing | Is it available when the need occurs? |
| Stigma | Does devaluation, exclusion or anticipated mistreatment obstruct its use? |
| Information visibility | Does accurate, usable information reach the person? |
| System continuity | Do responsibility, medication, information or support transfer successfully between services? |
These domains may overlap. They are proposed areas of inquiry, not five independent factors or a validated scoring scale.
Stigma belongs in this analysis because access can be obstructed by how a person is treated, not only by geography or capacity. The Public Health Agency of Canada's substance-use stigma primer describes discrimination, mistrust and barriers within healthcare, and supports examining institutional practices alongside language. [3] TRDP's hypothesis is that such processes can make a formally available protection practically unusable. Their role must be investigated in the specific setting.
Choosing privacy, declining an unwanted intervention or changing a treatment goal must not automatically be recorded as a person's failure. Consent and acceptable alternatives are part of the inquiry.
What the evidence supports—and what it does not
Research gives strong reasons to examine specific protective functions and transitions:
- Ongoing opioid agonist treatment: A systematic review found lower all-cause mortality during treatment than outside treatment in cohort studies. This is evidence about treatment in specified populations, not a test of a general Safety Continuity score. Observational comparisons remain vulnerable to differences between people and periods in and out of treatment. [4]
- Starting care and completing the next connection: A randomized trial found greater engagement in treatment at 30 days after emergency-department initiation of buprenorphine with coordinated follow-up than after referral or a brief intervention. The demonstrated outcome was engagement; it was not a mortality trial. [5]
- Preserving treatment through custody and release: A randomized US trial found greater return to community methadone treatment among participants assigned to continuation during incarceration than among those assigned forced withdrawal. It studied people who wanted to continue methadone; it does not justify incarceration as a route to care. [6]
There are also findings that constrain the argument. The HEALing Communities trial implemented a broad package of evidence-based practices across US communities but did not demonstrate a statistically significant reduction in opioid-related overdose mortality during its primary comparison period. [7] Delivering more activities, achieving engagement and reducing deaths are different outcomes.
None of these studies validates TRDP's complete framework. A transition can also coincide with changes in health, housing, drug supply, tolerance or exposure. An adverse outcome following discharge does not establish that the discharge or a failed handoff caused it.
Seven propositions worth testing
The existing conceptual manuscript sets out seven hypotheses. They remain questions for investigation:
| Proposition | Proposed test |
|---|---|
| P1 · Continuity gradient | Does a profile of interruptions improve prediction or decisions beyond established access and continuity measures? |
| P2 · Accumulating gaps | Do particular combinations of gaps disable complementary protections, and can acceptable substitutes change that relationship? |
| P3 · Stigma pathway | Does documented stigma precede changes in trust, treatment, delay or usable access that help explain a later outcome? |
| P4 · Transitions | Within a defined transition, do outcomes differ when protection is preserved, interrupted or replaced, after accounting for other relevant changes? |
| P5 · Lived experience | Does voluntary, structured inquiry identify actionable failures or solutions that routine records miss? |
| P6 · Wider measurement | Do nonfatal events, protective responses and access failures change conclusions that mortality data alone would support? |
| P7 · Information and access | Is understandable information more useful when the recommended protection is reachable and suitable? |
A finding that does not support a proposition must remain visible. More observations, moving stories or an attractive diagram cannot substitute for a valid comparison.
Follow one actual handoff
A useful first application is a bounded transition in which wanted protection can be checked before and after a change.
Consider a fictional example: a person leaves hospital expecting to continue a prescribed medication. A receiving pharmacy has been named, but its capacity and opening hours have not been confirmed. The paperwork is complete. Whether medication is accessible remains unknown.
An investigation could record:
- The need and timing: Which protective function is relevant, for whom, and during what period?
- The arrangement: What was offered, accepted and confirmed by the receiving service?
- Actual access: Was the function usable at the required time?
- Any interruption: When did it begin, how long did it last, and what evidence establishes that?
- Replacement or restoration: Did an acceptable alternative work, or was the original function restored?
- What followed: Record access, experience and health outcomes separately.
Use “unknown” when completion cannot be established. A referral is not proof of receipt; a dispensing record is not proof of ingestion. Record clinical changes, service inability, informed choices and unresolved reasons separately.
The first measures should remain a profile: completed handoffs, interruption duration, restoration time and practical usability. Combining them into one score could conceal a critical barrier. No clinical threshold or eligibility test has been established.
Whose knowledge counts
People who navigate these systems may know about an unanswered call, an inaccessible entrance, an unofficial workaround or a condition omitted from a service listing. TRDP calls the proposed use of that knowledge Lived Experience as System Intelligence.
That is a commitment to making experiential knowledge consequential in question-setting, interpretation and decisions. It is not a claim that one person's experience establishes population prevalence. Participation should be voluntary, supported and appropriately compensated, with control over personal information. People must remain eligible for support when they decline research participation.
Families and chosen supporters can contribute their own observations and needs without overriding another adult's choices. Indigenous-specific work requires the relevant communities' leadership and information-governance agreements.
What would count as progress?
A useful early result could be modest and concrete: a failed connection is documented, a workable repair is introduced, and completed access is checked again. That would be evidence about that repair in that setting.
The larger theory must earn its place through reliable measurement and comparisons with existing approaches. It should help identify something important, improve a decision, or make an overlooked interruption easier to repair.
Safety should survive the transition. Safety Continuity asks us to check whether it does.
Sources
- Haggerty JL, et al. Continuity of care: a multidisciplinary review. BMJ. 2003;327:1219–1221. Open article.
- Levesque JF, Harris MF, Russell G. Patient-centred access to health care: conceptualising access at the interface of health systems and populations. International Journal for Equity in Health. 2013;12:18. Open article.
- Public Health Agency of Canada. A Primer to Reduce Substance Use Stigma in the Canadian Health System. 2020. Official resource.
- Santo T Jr, et al. Association of Opioid Agonist Treatment With All-Cause Mortality and Specific Causes of Death Among People With Opioid Dependence: A Systematic Review and Meta-analysis. JAMA Psychiatry. 2021;78:979–993. Article and corrections.
- D'Onofrio G, et al. Emergency Department–Initiated Buprenorphine/Naloxone Treatment for Opioid Dependence: A Randomized Clinical Trial. JAMA. 2015;313:1636–1644. Study record.
- Rich JD, et al. Methadone continuation versus forced withdrawal on incarceration in a combined US prison and jail: a randomised, open-label trial. Lancet. 2015;386:350–359. Open article.
- HEALing Communities Study Consortium. Community-Based Cluster-Randomized Trial to Reduce Opioid Overdose Deaths. New England Journal of Medicine. 2024;391:989–1001. Study; NIH study report.
Research status: public edition adapted from TRDP's conceptual manuscript v1.4, Selective Emergency and Safety Continuity Revision 2, and the current scoping-review protocol. This is a proposed framework, not a peer-reviewed validation study. AI assisted preparation and source checking. Sources checked 4 November 2026.