Insights & biomarker education

What Insights covers

Insights is your understanding layer. It has tabs for your labs and your wearables, plus the cross-cutting reads that tie them together: your biological age, biomarker scorecards, habit signals, a weekly summary, and the records worth adding next. Everything is built only from data you have on file, and everything is educational, not a clinical judgment.

The screen now starts with a focused lab-review workspace. Accepted markers, pending review counts, and the next record question come first; search, education, wearables, panel-readiness, and advanced source context stay grouped so you can open the layer you need without scanning a wall of cards.

Whole-health synthesis readiness

The top of Insights now shows whether your record is ready for a whole-health Ojava Coach review. It checks records, labs, medications, wearable summaries, nutrition or body context, and clinical-context consent before staging a Coach prompt. If a piece is missing, the card names the next best input instead of pretending the review is complete. Source-acquisition planning uses the same gaps to name the most useful next source path, such as an existing lab PDF, a wearable export, a nutrition log, or an activation-gated records connector.

This does not create a second AI pipeline. The card is deterministic, and the button hands the same record-grounded context to Ojava Coach. The answer stays general wellness and visit prep: summarize what is supported, what is missing, and what to ask a clinician, never diagnose, prescribe, monitor emergencies, or decide care.

Evidence support map

When you ask what supports an answer or what to prioritize first, Ojava can map the reviewed whole-health context before Coach explains it. The map separates direct record support, partial support that still needs review, missing inputs, and self-tracked wellness context from wearables or nutrition. It names the next result or source worth reviewing without treating any single domain as the whole answer.

This is a support map only. It does not grade studies, calculate clinical certainty, prove causality, diagnose, rank treatments, order labs, prescribe, monitor emergencies, or decide care.

Your marker status at a glance

When you bring in lab results, Ojava bands each marker against a general reference range to show where it sits compared to typical values for adults. The grid displays markers grouped by system (Metabolic, Cardiovascular, Inflammation, Thyroid, Liver, Kidney, Hematology, Iron studies, Vitamins & minerals, and Electrolytes), with each marker shown as either in the typical range, above typical, or below typical. This is not a clinical judgment; it is an educational reference point to help you understand your own readings.

Coverage spans the markers people reasonably test today: a standard metabolic panel, a complete blood count with its indices, standard and advanced lipids, a thyroid panel, iron studies, common vitamins and minerals, and specialty markers across cardiac, clotting, urinalysis, pancreatic, endocrine, thyroid-antibody, autoimmune, immunology, allergy, fatty-acid, environmental exposure, specialty-screening, and genetic reports. AMH, thyroid antibody families, heavy-metal results, allergen IgE panels, fatty-acid details, tumor-marker-style labels, multi-cancer screening labels, and ApoE genotypes are recognized as report context and get plain-language education. Advanced heart-health panels are recognized too, including LDL particle number, small dense LDL, medium LDL, LDL peak size, HDL particle number, Lp-PLA2, myeloperoxidase, oxidized LDL, F2-isoprostanes, ADMA, SDMA, and TMAO. Markers whose typical range depends heavily on context, sex, age, reporting units, sample type, or the reason for testing are deliberately not banded, because a single general range would mislead more than it helps.

When you ask Coach about an advanced lipid, lipoprotein fractionation, or cardiology-marker panel, those same accepted rows can become a compact readiness packet. Coach can name which advanced cardiovascular markers are represented, which related marker families are still missing, and the education-only boundary. It does not calculate ASCVD, PREVENT, MESA, plaque, event, diagnosis, treatment, medication, lab-order, or treatment-risk outputs.

The grid shows only markers you actually have on file, updated from your newest reading per marker. Ojava never bands a value it suspects is in a different unit (for example, a glucose reading that looks like it arrived in mmol/L when the formula expects mg/dL), so implausible readings are silently skipped.

Panel catalog coverage

Ojava also keeps a panel-catalog view over the same accepted lab results. It can explain which marker families are represented, which named target markers are still missing from your record, and whether the coverage looks like core or add-on style context. The current targets cover 74 uniquely resolved Standard markers and 149 uniquely resolved Concierge markers (external reference-panel depth names, unrelated to Ojava’s own Free, Starter, and Pro plans), and the evals guard against alias double-counting. This is a readiness and education map only; it is not a lab order, checkout flow, diagnosis, treatment plan, or external score.

Before those accepted rows feed Coach, a lab draw reconciliation packet can also check whether panel rows still have the basics a large report needs: draw dates, source grouping, no same-draw duplicate marker rows, and no same-marker unit conflicts. That source-quality layer keeps the panel map easier to trust without turning it into lab validation or medical interpretation.

Per-system roll-ups and narratives

Below the grid, each system you have markers for gets a plain-language summary: what those markers generally reflect, how many of your readings in that system are in the typical range (for example, “3 of 4 in typical range”), and which markers are worth a look (any that fall outside the typical band). This is the Function Health pattern, purely a count; Ojava never assigns a score or grade.

Each system summary includes a general educational narrative about what those markers measure (for example, “These markers reflect how your liver is processing and clearing things from your body”) and a retest cadence that is commonly discussed for that system. The cadence is an educational note only, never a personal lab order.

Markers worth watching

Ojava highlights markers that are outside your typical range, or markers that are moving notably even if they are still in range. The “worth watching” list is ranked by how much each marker stands out: outside-typical markers rank highest, followed by markers with notable trends. It is a way to focus a conversation with your clinician on the readings that changed or shifted.

A trend is calculated from your earliest and latest readings. If you have only one reading per marker, no trend appears. If you have multiple readings the same day, Ojava averages them to show one point per day.

Biomarker education cards

Each marker you have gets an educational card. The card explains what that marker generally reflects (the reference range, not a personal target), things that commonly move the value, how often it is typically rechecked, and a few questions you could raise with your clinician about it. Cards appear only for markers you actually have on file.

The education follows the standard lab-result-interpretation pattern: meaning (what it measures), contributing factors (what affects it), retest cadence (how often it is typically revisited), and clinician questions (conversation starters). Nothing here is personalized; it is all general education so you can hold a more informed conversation with your doctor.

Trend charts and reference context

Where you have repeated measurements, Ojava shows a per-marker trend chart: your values over time, plotted against the general typical range band. The chart shows raw values (what you actually measured), not smoothed or interpolated. If you have measurements on the same day, they are averaged into one point.

The trend calculation is simple: oldest versus newest reading, as a percent change. A marker trending up or down, even if it is still in the typical range, is worth raising with your clinician because the direction can be more informative than a single snapshot. Ojava shows the trend direction (up, down, or flat) and the percent change since your first reading.

Biological age estimate

When you have enough of the right markers on file, Ojava can estimate your biological age using the PhenoAge formula. This is based on nine specific markers: glucose, hs-CRP, creatinine, albumin, alkaline phosphatase, white blood cells, lymphocyte percent, MCV, and RDW. If any of those nine are missing, Ojava shows you how many you still need to complete the picture instead of fabricating a number.

The biological age is not a diagnosis or a health score. It is an estimate based on a published research formula, applied to your actual markers, one data point among many. Ojava never invents a value; if the formula cannot run because markers are missing, it simply says so and asks you to add the remaining labs.

The result card now carries its research basis beside the estimate: the Levine PhenoAge equation, the DNAm PhenoAge lineage paper, and the 2026 Nature Medicine cohort paper linking PhenoAge-defined age gap with early-onset cancer risk at an association level. That last citation is education context only. Ojava does not calculate cancer risk, change screening eligibility, diagnose cancer, or turn biological age into a treatment target.

If you save an outside epigenetic-clock, pace-of-aging, appearance-age, or organ-system age result from the Body tab, Ojava keeps that as user-entered external-result context. Coach can reference the saved value only when you allow it, and it stays separate from the PhenoAge estimate Ojava computes from reviewed labs.

Biomarker scorecards

When the needed labs and body metrics are present, Ojava calculates research-backed metabolic, cardiometabolic, lipid-pattern, immune-pattern, kidney-pattern, and liver-pattern indices such as HOMA-IR, triglyceride-glucose index, METS-IR, cardiometabolic index, lipid accumulation product, visceral adiposity index, triglyceride-to-HDL ratio, atherogenic index of plasma, total cholesterol-to-HDL ratio, LDL-to-HDL ratio, non-HDL cholesterol, ApoB-to-ApoA1 ratio, remnant cholesterol, neutrophil-to-lymphocyte ratio, platelet-to-lymphocyte ratio, systemic immune-inflammation index, systemic inflammation response index, aggregate index of systemic inflammation, BUN-to-creatinine ratio, AST-to-ALT ratio, and albumin-to-globulin ratio. These scorecards are deterministic, fail closed when the required markers are missing or implausible, and use the same finished numbers that Ojava Coach reads. Where research formulas do not have one universal cutoff, Ojava shows them as trend context instead of assigning a verdict. Lipid-pattern, HOMA-IR, triglyceride-glucose index, cardiometabolic index, lipid accumulation product, visceral adiposity index, METS-IR, CBC-derived immune-pattern scorecards, and kidney/liver/protein organ-pattern scorecards also render linked source references beside the card when shown, drawing from the shared cholesterol, non-HDL, Matthews HOMA, Simental-Mendia TyG, Wakabayashi cardiometabolic index, Kahn lipid accumulation product, Amato visceral adiposity index, Bello-Chavolla METS-IR, Zahorec NLR, Kurtul platelet-to-lymphocyte ratio, Hu systemic immune-inflammation index, Qi systemic inflammation response index, Zinellu aggregate index, URMC BUN-to-creatinine, Botros and Sikaris De Ritis ratio, and MedlinePlus A/G ratio citation registry used by Ojava Coach.

For cardiology prep, accepted ApoB, Lp(a), and hs-CRP rows can also be grouped into a risk-enhancer discussion packet with source-backed thresholds from ACC/AHA materials. This is not an ASCVD, PREVENT, MESA, plaque, event, diagnosis, medication, lab-order, or treatment-risk score. It is a way to bring the right marker names, units, missing context, and citations into a visit-prep conversation. When ApoB and comparison lipids are present, the packet can also show ApoB beside LDL-C, calculated or direct non-HDL-C, and triglycerides, with a linked European Heart Journal discordance source explaining why those standard lipid values are not interchangeable proxies for ApoB. The marker education cards also show linked source references for matched markers, so ApoB, Lp(a), hs-CRP, and other explained labs do not rely on hidden citation metadata.

When you ask about cholesterol history, cumulative LDL exposure, or non-HDL exposure over time, Coach can use accepted dated lipid rows to calculate time-weighted LDL-C and non-HDL-C history in mg/dL-years. Non-HDL-C comes from a direct non-HDL result when present, or from total cholesterol minus HDL-C on the same draw date. The packet carries longitudinal LDL exposure source links, but it does not calculate ASCVD, PREVENT, MESA, plaque, event, or lifetime risk, set a target, recommend a statin, start treatment, order labs, or clear safety.

The same lab context layer can build a pre-lab window for Coach when an accepted result has a date. It summarizes nearby logged sleep, activity, meal or fasting timing, symptoms, and medication or supplement context so a visit-prep conversation can start with what was actually on file. It is not a lab-validity check or a cause finder.

For lipid panels, Coach can also prepare a fasting or nonfasting draw-context packet when you ask. It lines up accepted triglyceride, HDL, LDL, non-HDL, ApoB, or Lp(a) rows with nearby meal and fasting logs, cites the EAS/EFLM nonfasting lipid-profile consensus, and turns the result into questions for a clinician. It does not calculate ASCVD, PREVENT, MESA, plaque, or event risk, order repeat testing, set lipid targets, or give fasting instructions.

If your labs look reassuring on a portal, Coach can also prepare a typical-range debrief. It combines the general typical-range view, prior personal medians when enough same-unit history exists, missing context such as ApoB, Lp(a), or hs-CRP when lipid rows are present, and linked MedlinePlus plus cardiology-marker references. The debrief is for clinician questions only. It does not say a result is safe, calculate ASCVD or PREVENT risk, order tests, set targets, recommend imaging, or decide care.

The scorecards are educational indices only. They are meant to help you understand how related markers fit together before a visit, not to diagnose insulin resistance, diabetes, heart disease, infection, inflammatory disease, kidney disease, liver disease, visceral fat status, or insulin resistance, and not to change treatment.

If you ask Coach about MASLD, fatty liver, steatosis, liver enzymes, FibroScan, elastography, hepatic fat, fibrosis, or FIB-4, Ojava can prepare a separate liver-metabolic source packet from accepted rows. It brings liver enzymes, platelets, bilirubin, albumin, metabolic and body context, alcohol context, medication and supplement context, and imaging or report text into one source-backed clinician conversation prep view. It can also name whether age, AST, ALT, and platelet inputs are present for clinician FIB-4 discussion without calculating the score. The packet links AASLD nomenclature and practice-guidance sources, but it does not diagnose MASLD or MASH, stage fibrosis, calculate FIB-4, interpret elastography or imaging, prescribe diet, order labs, recommend treatment, or refer to hepatology.

If you ask about DEXA, bone density, T-score, fracture history, FRAX, fall prevention, walking steadiness, vitamin D, calcium, or PTH, Ojava can prepare a bone and fall-prevention source packet from accepted rows. It brings bone-density report language, fracture or fall history, mobility context, relevant labs, body context, medication and supplement context, and FRAX-input details into one clinician conversation prep view with linked BHOF, USPSTF, CDC STEADI, and FRAX sources. It does not diagnose osteoporosis or osteopenia, calculate FRAX, score fall risk, interpret imaging, recommend treatment, prescribe supplements or exercise, order labs or imaging, or refer to a specialist.

When enough reviewed inputs are present, the same scorecard also shows an allostatic-load context index across inflammation, resting heart rate, HDL, glycemic pattern, body-size context, blood pressure, and albumin, with linked sources for the compact allostatic-load marker-core research, the MacArthur multi-system framework, and scoring-method variability. It is an availability-limited trend-context count only, never a validated allostatic-load implementation, diagnosis, medical risk score, mortality predictor, or treatment plan.

Cardiovascular wellness score

Ojava can assemble an AHA Life’s Essential 8 style cardiovascular wellness score from source-backed components: diet quality, physical activity, nicotine exposure, sleep duration, BMI, non-HDL cholesterol, glucose, and blood pressure. The overall score appears only when all eight components are present. Diet quality and nicotine exposure can come from accepted, consented self-report observations, but not from generic notes or ordinary food logs. If either lifestyle input is missing, the signed-in Insights card can save an explicit reviewed check-in row for that LE8 input, then the same deterministic score path re-runs from the accepted observations. If those inputs are still missing, Ojava shows the completed components and the missing pieces instead of inventing an answer. HEI-2020 can appear as a separate diet-quality readiness result only when exact HEI constituent rows are reviewed; ordinary food logs remain nutrition context, not a substitute diet score. The signed-in card also shows linked research-basis sources beside the score so the AHA structure, activity guidance, and sleep-duration basis stay visible where the number appears.

This is also the current primary basis for the Ojava Health Score. Ojava treats a complete Life’s Essential 8 result as the headline 0 to 100 score, while PhenoAge and other published components can appear as companion context. If the validated headline score is incomplete, Ojava Coach receives a readiness note and missing-input list instead of a one-number score. When Coach receives the score packet, it also receives the configured source freshness windows and the reviewed source-date range so companion context stays visibly separate from the headline Life’s Essential 8 basis.

This is a wellness checklist, not a clinical risk calculator, insurance score, diagnosis, or treatment plan. It helps you see which everyday behaviors and measured factors are already on file, and which ones would make the picture more complete before a clinician conversation.

Wearable reads and trends

The wearables tab brings your daily physiology into the same understanding layer: readiness and recovery, heart-rate variability, resting heart rate, sleep, VO2max and fitness age, training load, and source-provided mobility signals such as walking steadiness and six-minute walk distance. It also shows source-provided blood-pressure summaries when your export includes paired systolic and diastolic readings, source-provided glucose summaries from CGM or glucose-app rows, source-provided oxygen and breathing summaries from SpO2 or respiratory-rate rows, source-provided sweat and electrolyte summaries from fluid-loss, sweat-rate, sodium, or electrolyte rows, source-provided body-composition summaries from smart-scale or health-app rows, source-provided arterial-stiffness and cardiovascular-age summaries from PWV or source age rows, plus vendor scores such as Body Battery, Training Readiness, sleep, Strain, recovery, stress, energy, or Movement Index when your export contains them. Each Ojava-computed value appears from your own data, and these are the same finished numbers Ojava Coach reads, so the chart and the conversation always agree. Vendor scores stay labeled as source scores instead of being re-derived. The full metric-by-metric detail lives on Wearables & longevity.

PWV and source cardiovascular-age rows stay as source-provided trend context. Insights can show the latest source value beside the user’s own recent logged average and linked PWV research sources, but it does not calculate cardiovascular risk, diagnose, monitor, alert, recommend treatment, or decide care from those rows.

Habit signals

Ojava looks for associations between the habits you log (workouts, fasting, wellness sessions, and the like) and your outcomes (sleep, recovery, readiness), and surfaces the ones that stand out, for example, that your readiness tends to be higher on days after an easy training day. These areassociations, never causation: they are a prompt to notice a pattern in your own data, not proof that one thing caused another.

Cross-domain lifestyle discovery uses the same association engine across user-authored lifestyle context and source-backed outcomes after both sides have enough observed days. The output stays a private observation for Coach, not diagnosis, treatment, medication dosing, medication start or stop advice, or behavior scoring.

Nutrition targets

When you have logged your body metrics, Insights shows a nutrition-targets card: your estimated daily energy needs (BMR and TDEE), your BMI, body-shape indices from waist, height, and BMI, and a maintenance macro split. These are the same figures Ojava Coach uses, derived identically, so the card and the chat never disagree. The full nutrition surface is on Nutrition, food & body.

Your weekly summary

A plain-language week-in-review compares this week to last across your key signals (sleep, activity, readiness, and trends) and highlights the biggest changes, so you can see the direction of travel without reading every chart.

Records worth adding

Insights also shows a record-repair checklist: common markers that are not on file yet, or results whose newest entry is older than a year. Each item explains what is missing or older, gives a clinician question, and links back to Add records so you can attach an existing result if you have one. This is never a recommendation to order a specific test; it is a prompt to improve your record or prepare a question before your next visit.

Lab search and filtering

The screen includes a search box so you can quickly find a marker you are interested in, or filter by category (Metabolic, Cardiovascular, and so on). You can also see which categories have the most data and jump between them.