Why Build a Reference Profile?
Regional physiology values are often assembled from different measurements, study populations, and experimental conditions. A table can look complete while combining, for example, a fasting pH value, a fed-state bile-salt concentration, and a transit estimate defined for a different region. Such a table is not necessarily a coherent description of any one reference condition.
The purpose of an audit is not to declare one set of values universally correct. It is to make the intended reference condition explicit, preserve the evidence and units behind each value, and identify entries that need checking before they are used. The earlier material on regional GI compartments, physiology measurements and their biases, and covariation among GI physiology inputs provides the relevant context: boundaries and measurement definitions matter, and inputs may vary together.
Specify the Profile Before Filling It
A workable profile needs an explicit scope. For the examples below, suppose the goal is a hypothetical fasted-adult reference profile. This phrase is a scenario label, not evidence that the listed values are universal fasting means. The table uses five broad regions—stomach, duodenum, jejunum, ileum, and colon—and records pH, mean residence time, available luminal fluid volume, and bile-salt concentration.
For each quantity, record what the value means. A transit-time entry should specify whether it is a mean, median, or another summary, and how the region and interval were defined. A volume entry should distinguish an estimate of available luminal water from a total secreted volume or cumulative fluid throughput. A bile-salt entry should state whether it is a concentration, the analyte definition, and the sampling condition. A pH entry should be associated with a region and condition rather than treated as a permanent property of that region.
The following numbers are deliberately hypothetical. They illustrate table construction and arithmetic, not a recommended parameter set or measured human dataset.
| Region | pH | Mean residence time (h) | Available fluid volume (mL) | Bile salts (mM) |
|---|---|---|---|---|
| Stomach | 2.0 | 0.50 | 50 | 0.05 |
| Duodenum | 6.0 | 0.25 | 30 | 3.0 |
| Jejunum | 6.5 | 1.5 | 100 | 2.0 |
| Ileum | 7.0 | 3.0 | 100 | 1.0 |
| Colon | 6.5 | 20 | 50 | 0.10 |
The table is easier to audit when paired with a source record. At minimum, retain the citation or dataset identifier, the population and condition, the original region definition, the original reported value and unit, and the rule used to select or transform the value. If a value is an estimate chosen from a range rather than a direct measurement, say so. Keep the original entry as well as the harmonized entry; otherwise, a unit conversion or regional mapping cannot be independently checked.
Screen Values Without Mistaking Screens for Laws
A first-pass plausibility screen asks whether each value is in a broad, credible range for its named quantity and region. Screening envelopes can help catch transcription errors, unit mismatches, and region-label problems. They are not universal physiological bounds: a value outside an envelope may reflect a real condition, a different measurement definition, or an unsuitable envelope. It should prompt investigation rather than automatic replacement.
For illustration, an audit team might use the following deliberately broad envelopes as flags while reviewing a proposed fasted-adult table. They are not supplied as authoritative reference ranges; a real analysis should document and justify its own source-supported screens.
| Region | Illustrative pH screen | Illustrative time screen (h) | Illustrative volume screen (mL) | Illustrative bile-salt screen (mM) |
|---|---|---|---|---|
| Stomach | 1–5 | 0.1–4 | 10–300 | 0–0.5 |
| Duodenum | 4.5–7.5 | 0.05–1 | 5–100 | 0.1–15 |
| Jejunum | 5.5–7.5 | 0.5–4 | 20–300 | 0.05–10 |
| Ileum | 6–8 | 1–8 | 20–300 | 0.01–5 |
| Colon | 5–7.5 | 8–48 | 5–150 | 0–1 |
The illustrative profile falls inside these example screens. That result only says that these broad checks do not flag the displayed entries. It does not validate their provenance, establish precision, or show that all four values in a row were observed under the same conditions.
Worked Example: Harmonizing a Volume Unit
A hypothetical source reports a jejunal available-fluid estimate of 0.12 L. The draft table uses mL. Convert the source value and compare it with the illustrative jejunal screen of 20–300 mL.
Use \(1\ \mathrm{L}=1000\ \mathrm{mL}\). Therefore,
The converted entry is 120 mL, which lies within the illustrative screen. The conversion is dimensionally consistent because liters cancel and the result is in mL. The audit record should preserve both “0.12 L” as reported and “120 mL” as the harmonized value, together with the conversion factor. The screen does not establish that the source’s “available fluid” definition matches the definition used elsewhere in the table.
Check Internal Consistency and Dependencies
After checking individual values, compare the entries with one another. Internal consistency is not the same as requiring every value to follow a simple monotone pattern. Instead, ask whether the region labels, condition, units, and definitions can plausibly describe one stated scenario. A profile may combine values from different sources, but differences in population, fasting or meal state, sampling time, and measurement method should be visible and assessed.
Some arithmetic checks are useful if their assumptions are met. If mean residence times refer to consecutive regions on the same pathway, under compatible conditions and with compatible definitions, their sum gives the mean total residence time across those regions. For the illustrative table:
This is a conditional consistency calculation, not proof that the component estimates are comparable. It would be misleading if one entry were a median, another a mean, or the estimates came from incompatible populations or protocols. Nor does the sum imply that every person’s total residence time is 25.25 h.
Concentration and volume can also be combined to calculate the amount present in a region at a stated time, provided they refer to the same region and compatible conditions. Since \(1\ \mathrm{mM}=1\ \mathrm{\mu mol/mL}\), their product has units of micromoles:
This gives a regional inventory estimate, not a secretion rate, absorption rate, or cumulative amount passing through the region. Those quantities require additional information.
Worked Example: Calculating a Regional Bile-Salt Inventory
In the hypothetical duodenal row, bile-salt concentration is 3.0 mM and available fluid volume is 30 mL. Assuming both values describe the same reference condition, calculate the implied inventory.
Because \(3.0\ \mathrm{mM}=3.0\ \mathrm{\mu mol/mL}\),
The implied amount is 90 μmol. This arithmetic is a useful cross-check: if a source or downstream table reports a different inventory for the same concentration and volume, the units or values need review. It does not establish that 90 μmol is a measured amount, because the calculation inherits the assumptions and uncertainty of both input estimates.
Covariation matters during profile assembly. Inputs may change together with fasting or meal state, and measurements may be linked by their shared protocol or population. Combining separate marginal averages can create a synthetic profile that does not represent a coherent person or condition. Preserve the condition labels and source relationships; do not manufacture a “typical” joint profile by independently choosing a central value for every column without checking whether those choices are compatible.
Worked Example: Finding a Unit and Condition Mismatch
A hypothetical draft lists ileal mean residence time as 180 h. The source record says “ileal transit: 180 min,” while the other entries are reported in hours. The same draft labels the entire table fasted, but its gastric pH source describes a fed condition. Identify the audit findings and correct what can be corrected.
Time conversion. Since \(60\ \mathrm{min}=1\ \mathrm{h}\),
The draft’s 180 h is a unit-transcription error; the harmonized entry is 3.0 h. It falls within the illustrative ileal time screen of 1–8 h, though that alone does not validate the source estimate.
Condition mismatch. The gastric pH value cannot be made fasted merely by changing its label. The source describes a fed condition, so the audit should flag the mismatch and either locate a compatible fasted estimate, redefine the entire profile’s condition if appropriate, or retain the entry as an explicitly qualified exception. The available information does not justify calculating a replacement pH.
A Practical Audit Sequence
A compact sequence helps separate clerical problems from scientific disagreements. The essential output is an auditable decision for each entry: retain, convert, qualify, replace with a documented alternative, or leave unresolved.
State the reference population, physiological condition, region scheme, and summary type intended for the profile.
Record the source, original definition, condition, value, and unit. Keep source values separate from harmonized values.
Convert units with explicit factors and document how source regions map to the profile’s regions. Do not imply greater precision than the source supports.
Apply documented plausibility flags, then check cross-entry definitions, conditions, and any arithmetic identities that actually apply.
For every flagged value, note the reason, resolution, remaining uncertainty, and any effect on the intended use.
Alternatives, Diagnostics, and Limitations
A single reference profile is compact and convenient for a baseline model specification, comparison, or reproducible calculation. Its main weakness is compression: it hides between-person variation, within-person changes, and uncertainty in the underlying measurements. An alternative is to retain ranges or distributions for uncertain inputs, or to define multiple coherent scenarios such as fasting and fed profiles. That approach represents variability more directly but needs more evidence and careful coordination among inputs. This tutorial concerns assembling and auditing the profile; it does not specify a gut absorption model or a procedure for propagating these inputs through one.
Useful diagnostics include values far outside a documented screen, abrupt changes that prompt a region-label check, totals that disagree with source definitions, suspiciously identical values copied across regions, and unit patterns that differ from the rest of a column. A flagged pattern is a reason to inspect provenance—not a reason to “smooth” the data until it looks tidy. Likewise, a table that passes every numerical screen may still be unreliable if its sources describe incompatible conditions.
- Do not treat a screen as a physiological law. The illustrative envelopes above are prompts for review, not universal limits.
- Do not silently mix conditions. A value from a fed protocol remains fed-context evidence even if placed in a fasted table.
- Do not confuse concentration with amount. A concentration requires a compatible volume to calculate an inventory, and an inventory is not a flux.
- Do not overstate precision. A converted value can have more decimal places mathematically than the source warrants biologically.
Check Your Understanding
Answer each question, then compare with the brief solutions below.
- A source reports 0.08 L of available fluid for a region. What value belongs in a table expressed in mL?
- A profile lists 2.0 mM bile salts and 40 mL fluid for the same region and condition. What inventory follows from these values, and what does it not represent?
- Why is it not sufficient to place a fed-state pH value in a fasted profile and label the row “fasted”?
- Under what conditions can regional mean residence times be summed to obtain a total mean residence time?
- A table passes all broad plausibility screens. Does that establish that its values come from a coherent reference population and condition? Explain.
Solutions. (1) \(0.08\times1000=80\) mL. (2) \(2.0\ \mathrm{\mu mol/mL}\times40\ \mathrm{mL}=80\ \mathrm{\mu mol}\); this is an implied regional inventory, not a rate or cumulative throughput. (3) The source condition remains fed; relabeling does not convert the measurement into fasting evidence. The entry needs a compatible source or an explicit qualification. (4) The times must refer to consecutive regions on the same pathway and have compatible definitions, conditions, and summaries. (5) No. Numerical screening cannot establish provenance or cross-source compatibility; those require separate checks of population, condition, region definitions, and measurement methods.