Health Span Calculator — Methodology
Last reviewed: 2026-11-09 · Model version v1.0.0
Scope
The health span calculator is decision support for senior-dog daily routines. Its explicit non-goals: individual lifespan prediction, diagnosis, treatment advice, and product recommendations. This page is the audit trail for every weight the model uses — any change requires a version bump and a changelog entry below.
Try it here: Health Span Calculator.
Inputs
- Age — years, 0.5–30, step 0.5.
- Breed & size class — searchable over a curated 81-breed baseline, plus a "Mixed / Rescue (pick size class)" entry; free text that doesn't match degrades to size class. Size classes: S (<25 lb) / M (25–50) / L (50–90) / G (>90 lb) — the class selector is hidden when a matched breed supplies it.
- Body condition (visual proxy) — ideal / under-condition / over-condition.
- Exercise — daily / a few times a week / rarely.
- Diet — complete & balanced fed by measure / mixed / unstructured.
- Dental care — daily / occasional / none.
- Vet cadence — twice a year / once a year / only when sick.
Scoring model (v1.0.0)
Score = 100 − Σ penalties, clamped by construction — maximum total penalty 70, so the floor is 30.
| Factor | Value | Penalty | One-line rationale |
|---|---|---|---|
| Body condition | over-condition | 18 | The routine factor most consistently associated with reduced lifespan and comfort (lean-fed dogs lived longer in Kealy et al. 2002). |
| Body condition | under-condition | 10 | In seniors, under-condition warrants vet evaluation first — it can signal disease. |
| Exercise | rarely | 12 | Mobility, muscle mass, joint and cognitive health associations. |
| Exercise | a few times a week | 6 | Partial coverage; consistency typically beats intensity. |
| Diet | unstructured | 10 | Complete-and-balanced, measured feeding; treats kept small. |
| Diet | mixed | 5 | Balanced base but unmeasured treat load. |
| Dental care | none | 14 | Periodontal disease is highly prevalent and painful even when dogs keep eating (AVDC). |
| Dental care | occasional | 7 | Partial coverage; daily brushing is the practical gold standard. |
| Vet cadence | only when sick | 16 | Misses early detection; senior wellness visits are typically twice-yearly (AAHA). |
| Vet cadence | once a year | 8 | Annual is standard adult cadence; senior cadence is higher. |
| All factors | best value | 0 | — |
Score bands
Bands use decision-support wording, never verdicts:
| Score | Band |
|---|---|
| 85–100 | Strong routine |
| 70–84 | Good, with gaps |
| 50–69 | Needs attention |
| < 50 | Worth a vet conversation |
Top-3 actions: each sub-best factor maps to one action line (comfort / monitoring / routine framing only; no treatment claims). Sub-best factors are sorted by penalty and the top 3 are rendered; every abnormal-sign path routes to the vet first. Zero penalties produces a praise line plus a keep-the-rhythm note.
Senior age thresholds
Size-class model: S 11 / M 9 / L 8 / G 7 years, with curated breed overrides in the breed dataset (leading giants earlier — e.g. Great Dane 6). This follows the inverse size–seniority convention in AAHA life-stage framing. Copy rule: "typically considered senior around age N".
Lifespan ranges
Per-breed published ranges plus size-class defaults (S 12–16 / M 10–14 / L 9–13 / G 6–10 years).
Hard rule: ranges never adjust with lifestyle inputs. Display copy: "Dogs this size typically live X–Y years" plus a standing "Published-range context — not a prediction for your individual dog" line.
Sources
Every citation is checked against the original document before it is displayed on-page.
| Source | Supports |
|---|---|
| Kealy RD et al., "Effects of diet restriction on life span and age-related changes in dogs", JAVMA 2002;220(9):1315–1320 | Body-condition weight rationale |
| AAHA Canine Life Stage Guidelines, 2019 | Senior threshold + twice-yearly cadence |
| AVDC owner resources — periodontal disease | Dental factor rationale |
| WSAVA Global Nutrition Toolkit | Diet factor framing |
Guardrails
- "Typically / associated with" fixed phrasing across all widget copy.
- No disease-risk percentages; no treatment claims; no product recommendations.
- A disclaimer always appears under results and inside the printable summary; "not a prediction" is shown twice.
- No
aggregateRatingor fabricated review schema.
Version history
- v1.0.0 (2026-11-02) — initial model: 5 factors / 11 penalty states, bands, senior thresholds, lifespan ranges, guardrails.
- 2026-11-09 — editorial pass: presentation only, no model change.