Where costs go from here
Into the 2027 bid cycle, plans face flat-to-negative effective rates while the specialty-access gap that drives avoidable admissions stays open — and, on current evidence, cannot be closed by adding specialist headcount. Our forward view: plans that hold the status quo will see avoidable-admission cost drift upward as prior-authorization friction lengthens time-to-specialist; plans that close the gap with need-based virtual specialty follow-up can bend the readmission curve on the conditions that dominate inpatient spend. The evidence below sizes the mechanism; the model at the end turns it into a number you can run on your own population.
Specialist access is slow — and supply isn't the lever
In a 2019–2023 analysis of roughly 114,000 commercially insured US patients, the average wait for a new neurology visit after a primary-care or ED visit was about 50 days (median 25). The finding that matters strategically: neurologist density was not associated with shorter waits. More specialists in a market did not mean faster access. Waiting is a delivery-model problem, not a headcount problem — which is why simply "adding specialists" hasn't worked and why a virtual-first model that reallocates access is the more plausible fix.
Neurology is used here as the best-measured proxy for ambulatory specialty wait times; the access dynamic generalizes to the cardiometabolic and pulmonary specialties that drive MA spend.
Longer waits produce worse outcomes in exactly the patients MA can't afford to lose
A review of Veterans Health Administration wait-time data found that longer waits led to small but statistically significant drops in outpatient utilization and to poorer health in elderly and vulnerable patients — including higher mortality, more preventable hospitalizations, and worse HbA1c control. Delay doesn't remove demand; it converts a manageable outpatient problem into a downstream inpatient one. For a chronic, dual-eligible-heavy MA population, that conversion is precisely where total cost of care escapes.
Post-discharge specialist contact reduces the readmissions that dominate spend
- In a COPD cohort, patients who did not attend a pulmonologist follow-up within 30 days of discharge had roughly 2.9× the odds of readmission within 90 days (odds ratio 2.91; 95% CI 1.06–8.01).
- In a ~94,000-patient Ontario study using trial-emulation methods, a physician follow-up visit reduced COPD-specific readmission, ED use, or death (hazard ratio as low as ~0.89) — though the all-cause effect was roughly null, and no single "optimal" follow-up day existed. The benefit came from access matched to medical need, not from a fixed calendar rule.
- Context for scale: COPD carries one of the highest 30-day readmission rates in medicine (~20%), yet in that population only 7.7% saw a pulmonologist within 30 days of discharge. The gap between evidence and delivery is enormous — and it is addressable.
Our program data tracks the literature
Across the Medicare Advantage and risk-bearing-provider populations we manage — client identities withheld — the pattern above holds: compressing time-to-specialist and delivering structured, need-based specialist follow-up is associated with fewer avoidable admissions in cardiometabolic and pulmonary cohorts.
| Program signal (observed) | What to insert (from your data, anonymized) |
|---|---|
| Time-to-specialist | [from ___ days to 24–48h] in a managed MA cardiometabolic cohort (N=___) |
| Readmission / avoidable admission change | [___% reduction], measurement window [___], comparison group [___] |
| Engagement | [___% six-month engagement] |
Every figure in this section must be an observed result published with its population, baseline, comparison group, and measurement window before external use. Insert real, de-identified numbers; do not generalize a single client's result. (This mirrors the claims discipline used across Circadian materials: Observed / Modeled / Projected / Validated.)
Two paths for total cost of care
Status quo. Flat-to-negative effective rates after coding-intensity recalibration, plus rising cardiology/pulmonology prior-auth denials, lengthen time-to-specialist. Finding 2 says that delay raises preventable admissions; Finding 1 says added supply won't shorten the wait. Net direction into 2027: avoidable-admission cost drifts upward, and the gap compounds because the usual fix (more specialists) doesn't work.
Close the gap. Need-based virtual specialty access and structured post-discharge follow-up attack the readmission driver directly (Finding 3). Direction: the readmission curve bends, with magnitude set by population size, baseline readmission rate, and cost per admission.
A model you can run on your population
| Input | Value | Source |
|---|---|---|
| Members with target conditions (CHF / COPD / diabetes) | [N] | Your population |
| Baseline 30-day readmission rate | ~20% (COPD) | Published (PLOS 2024) |
| Cost per readmission | [$ your plan average] | Your claims |
| Modeled risk reduction from need-based specialty follow-up | [10–25%, conservative] | Modeled from OR 2.91 / HR ≈0.89 |
| Projected avoidable readmissions averted | N × rate × reduction | Modeled |
| Projected gross savings | averted × cost per readmission | Modeled |
Projected, not guaranteed. This is an illustrative model, not a validated result. The literature-derived parameters are shown; the plan-specific inputs are placeholders. The published effect sizes come from observational cohorts (heterogeneous populations, no fixed optimal timing, all-cause effects weaker than condition-specific), so the risk-reduction input should be set conservatively and, ideally, validated against your own observed data before it is presented as anything more than a forecast.
What this forecast does and doesn't claim
- The supporting studies are observational (including one trial-emulation design); they establish association and plausible mechanism, not guaranteed causal savings for any single plan.
- The COPD timing study found no fixed optimal follow-up window and a roughly null all-cause effect — we intentionally frame the intervention as need-based access, not a scheduling mandate.
- Wait-time evidence is drawn from neurology and VA settings as the best-measured proxies; generalization to MA cardiometabolic populations is reasoned, not directly measured.
- Our own program figures are observed, not randomized, and must carry their methodology when cited.
Evidence base
Retrieved via PubMed. Please cite PubMed and the DOIs below when reusing this material.
- Laffargue EK, Van Der Goes DN, Wilson AM, Parziale SD, Sico JJ, Ney J. Neurology wait times after primary care or emergency department visits among the commercially insured US population: 2019–2023. Neurology. 2026;106(10):e218008. https://doi.org/10.1212/WNL.0000000000218008
- Pizer SD, Prentice JC. What are the consequences of waiting for health care in the veteran population? J Gen Intern Med. 2011;26(Suppl 2):676–682. https://doi.org/10.1007/s11606-011-1819-1
- Gavish R, Levy A, Dekel OK, Karp E, Maimon N. The association between hospital readmission and pulmonologist follow-up visits in patients with COPD. Chest. 2015;148(2):375–381. https://doi.org/10.1378/chest.14-1453
- Timing of follow-up visits after hospital discharge for COPD: application of a new method. PLOS ONE. 2024;19(7):e0302681. https://doi.org/10.1371/journal.pone.0302681
- Supporting — intervention effect size: Inglis SC, et al. Structured telephone support or non-invasive telemonitoring for heart failure. Cochrane Database Syst Rev. 2015;(10):CD007228. https://doi.org/10.1002/14651858.CD007228.pub3