The single biggest win here is energy: nearly every cause on the list drags it, and naming the cause lifts the floor of your day inside weeks — usually before the weight has moved at all. Sleep often comes back with it. Mood lifts the moment I lack willpower gets replaced with a name and a treatment. The cost is a doctor's visit and a few standard tests; the effort is one honest week of tracking. You will not always find a medical answer — but you cannot tell from the outside, which is the whole point of the workup.
What can actually be doing this
The body you wake up in is the output of many systems — thyroid hormone, cortisol, oestrogen and testosterone, insulin, sleep, appetite hormones, and every prescription on your shelf. Any of them can drift, and when one does, the weight follows. The differential a competent clinician works from is short, named, and treatable.
Thyroid running slow. A low level of thyroid hormone cuts the rate at which your body burns energy at rest by anywhere from fifteen to forty percent at clinically overt disease, and adds a puffy fluid retention older doctors still call myxoedema. About three in a thousand US adults are clinically low; another roughly four in a hundred sit in a milder grey zone, with women and older adults overrepresented. The cheapest single test of the whole workup — a thyroid stimulating hormone blood draw, the TSH — rules it in or out Hollowell et al. 2002Garber et al. 2012.
A sleep disorder hiding behind snoring. Obstructive sleep apnea — your airway collapsing dozens of times a night, fragmenting sleep and starving the blood of oxygen — both causes weight gain and is caused by it. The nightly oxygen dips and adrenaline surges raise stress hormones and insulin resistance; the broken sleep itself flips the appetite hormones the wrong way and produces measurable next-day overeating in laboratory studies. About one in eight men and one in seventeen women aged 30–70 meet criteria for at least moderate disease, and roughly four in five cases are undiagnosed Peppard et al. 2013Spiegel et al. 2004.
A medication on the list nobody summed. A 2015 systematic review catalogued nine drug classes that reliably push the scale up; the worst routine offenders are a handful of second-generation antipsychotics (olanzapine, clozapine), several antidepressants (mirtazapine, paroxetine, amitriptyline), corticosteroids like prednisone, insulin and the older sulfonylureas for diabetes, the non-vasodilating beta-blockers (atenolol, metoprolol), the seizure drugs valproate and gabapentin and pregabalin, and lithium. Most adults take at least one prescription drug, often more, and nobody has ever totalled the obesogenic load Domecq et al. 2015.
Cortisol gone wrong. Cushing's syndrome — too much cortisol, either made by the body or taken as medicine — produces a recognisable pattern: weight piles onto the trunk and the back of the neck, the face rounds, the legs and arms thin out, the skin bruises easily, blood pressure climbs. The version made by the body is rare. The version caused by a course of prednisone for asthma or by a steroid inhaler used too long is common and routinely missed, because the cause is the patient's own bottle Nieman et al. 2008.
Insulin resistance with the wrong hormone profile. Polycystic ovary syndrome runs at eight to thirteen percent of women of reproductive age, and is driven centrally by the body's cells responding poorly to insulin. The pattern is abdominal weight that resists the usual eat-less-move-more approaches, often with cycle irregularity, acne, or unwanted hair growth that the patient has lived with so long she no longer mentions it at appointments Teede et al. 2018.
The body the calendar is rearranging. A long-running cohort study of women through the menopause transition found that total weight rises only modestly across the change — but fat mass, especially around the abdomen, climbs sharply while muscle falls. The headline "menopause makes you gain" conflates a real composition shift with a scale shift; the trousers fit differently because the body is being remodelled, not because the calories changed Greendale et al. 2019Davis et al. 2012.
Weight that came back after weight that went. Anyone who lost a meaningful amount in the prior decade is carrying persistent metabolic and hormonal adaptation: a resting energy burn below what their body composition predicts, hunger hormones turned up, fullness signals turned down, all of it physiologically opposing the maintenance. Six years out from a televised weight-loss contest, the contestants were still running about five hundred fewer kilocalories per day of resting metabolism than their bodies predicted. The willpower theory of regain is wrong; the body is fighting back, not the person Fothergill et al. 2016Sumithran et al. 2011.
Short sleep on its own. Habitually getting six hours or fewer roughly halves the fullness signal in the blood and pushes the hunger signal up enough to measure on lab tests, and it tilts the brain's reward circuits toward palatable food. A pooled analysis of thirty studies found short sleepers significantly more likely to put on weight over time Cappuccio et al. 2008Spiegel et al. 2004.
The two-way loop with depression. Depression raises the risk of becoming obese (odds ratio about 1.58); obesity raises the risk of becoming depressed (odds ratio about 1.55) — no clean primacy. Lower activity, altered stress hormones, broken sleep, and the obesogenic class of antidepressants often prescribed to treat it all stack Luppino et al. 2010.
Caloric drift you cannot see. The gold-standard way to measure how much someone is actually eating — having them drink water tagged with traceable atoms and measuring what comes out — shows that adults with obesity underreport their food intake by about half and overreport their activity by about half, and not consciously. The commute that subtracted a walk; the coffee order that quietly doubled in calories; the after-school snack pattern around the kitchen at 4pm — each is small, and the sum is the gain. The arithmetic of energy balance still holds: a sustained surplus of about twenty kilocalories a day predicts a kilogram of weight a year for an average adult. The honest one-week food log is part of the workup, not a substitute for it Lichtman et al. 1992Hall et al. 2011.
What the workup actually finds
Nobody has run the single study you would want — a prospective cohort of "unexplained weight gain" patients all worked up the same way and the diagnostic yield reported by cause. The numbers that do exist are per-axis.
- Clinical low thyroid is in about three of every thousand US adults; the milder grey-zone version is in about one in twenty-three. Both lean female and older. Pretest probability rises sharply when fatigue, cold intolerance, or constipation accompany the gain, which is why the thyroid blood test is the universal first-line check Hollowell et al. 2002Garber et al. 2012.
- About one in eight men and one in seventeen women aged 30–70 have at least moderate sleep apnea, and roughly four in five are undiagnosed. A validated screening questionnaire plus a home sleep study finds it cheaply Peppard et al. 2013.
- The medication audit is the highest-yield single step in many real workups; the catalogue of obesogenic drug classes is broad, and most readers have never gone through it line by line Domecq et al. 2015.
- Polycystic ovary syndrome covers eight to thirteen percent of reproductive-age women, with a well-defined diagnostic pathway: history, androgen blood tests, and a pelvic ultrasound Teede et al. 2018.
- Cushing's is rare in absolute terms, but its three first-line screening tests are good and the cost of missing it is severe Nieman et al. 2008.
The pattern across causes is the one that matters: identifying and treating the cause restores energy, sleep, mood, and cardiometabolic risk before it restores all the weight. Thyroid replacement returns a few kilograms, mostly water. The mask-and-machine treatment for sleep apnea lifts daytime sleepiness and blood pressure but does not reliably move the scale without a parallel intake change. Switching from olanzapine to a weight-neutral antipsychotic recovers a substantial share of the gain over months. Tapering exogenous steroids reverses the Cushing's pattern over months. The reframing alone — the body is not betraying you; a discoverable cause is — moves a patient from self-blame back into agency Karmisholt et al. 2011.
What ignoring it costs over time
Each cause has its own consequence stream, and they accrue silently while the workup is being avoided.
Within months, untreated low thyroid becomes its own disease. The fatigue and the cold deepen. The cognitive slowing gets noticed by the people around you before you notice it yourself: the meeting you used to lead is led by someone else; the conversation at dinner is harder to follow. Depression often arrives by the time the diagnosis does, and it lifts unevenly even on replacement Garber et al. 2012.
Within years, untreated moderate sleep apnea quietly raises the odds of hypertension, an irregular heartbeat, stroke, and early death. The partner who first slept in the other room because of the snoring eventually sleeps there because of the morning irritability. The drives home from work at 5pm become the kind of drives that are almost-accidents you do not tell anyone about Marcus et al. 2008Peppard et al. 2013.
A steroid-driven Cushing's pattern, if missed, produces brittle bones, diabetes, and a weakened immune system inside a few years. The patient breaks a wrist on a low fall and the surgeon is the one who finally asks why an otherwise-healthy fifty-year-old has bone density in the bottom percentile Nieman et al. 2008.
Antipsychotic-driven gain destabilises the cardiometabolic baseline of patients who are already at higher cardiovascular risk from the underlying psychiatric condition. The 35-year-old on olanzapine for two years has the metabolic profile of a sedentary 55-year-old; the years lost to that compound forward Domecq et al. 2015Allison et al. 1999.
The body-composition shift at menopause raises diabetes risk and contributes to the post-50 jump in heart disease in women even when the scale barely moves. The version of the problem that "looks like nothing" on the scale is the version that lands hardest at the cardiology clinic a decade later Greendale et al. 2019Schwingshackl et al. 2017.
And the loop with depression closes itself. The self-blame about the weight feeds the depression that fed the weight; the medicines that treat the depression add to the weight; the relationship strain and the cancelled social calendar produce more depression. Each year the loop runs, the next year is harder to start Luppino et al. 2010.
How to actually run the workup
The single best move you can make today is to walk into a primary-care visit with an organised ask, not an open-ended complaint. The workup is mostly cheap, mostly covered, and finite.
Perimenopause changes the question. If your cycles are shortening, lengthening, or skipping, the body-composition shift is part of the differential. The strategy reorients from "lose weight" to preserve muscle and prevent abdominal accumulation: more protein, more resistance training, less reliance on long cardio sessions for body composition. The scale may not move much; the trousers will fit again because the composition under them is what changed Greendale et al. 2019.
For men, low testosterone belongs on the list. It presents as fatigue, low libido, erectile difficulty, flat mood, and slow weight gain — and diagnosis requires a morning total-testosterone draw on two separate mornings, not a single afternoon value. Replacement, when indicated, restores energy and lean mass; it is not a fat-loss drug, but the body composition tilts in the right direction.
What most stories about it get wrong
The four stories you have heard about your own weight that the evidence does not support:
- "My metabolism crashed at forty." Doubly-labelled water — again, the gold standard — shows that resting energy burn is roughly flat from age 20 to 60, then declines about seven-tenths of a percent a year. The mid-life cliff most adults believe in is not in the measurements. What did change is activity and muscle mass — both of which you can address; neither of which is a metabolic conspiracy against you.
- "It must be my thyroid." Low thyroid is real and worth ruling out, but it explains a minority of unexplained gain even in symptomatic groups, and the weight it accounts for is a few kilograms, mostly water. A normal thyroid blood test rules it out as the dominant cause; do not chase it twice Karmisholt et al. 2011.
- "Menopause makes you fat." The transition redistributes fat to the abdomen and reduces muscle without large total-weight change. Conflating the redistribution with absolute gain misframes the problem — and the right intervention Greendale et al. 2019Davis et al. 2012.
- "My last diet failed because I lacked discipline." Persistent metabolic adaptation after weight loss runs hundreds of kilocalories per day below predicted for years; hormonal hunger drive stays elevated across the same window. The physiology was actively opposing the maintenance. The reading of yourself you have been carrying is wrong Fothergill et al. 2016Sumithran et al. 2011.
The fifth and most demoralising story — I am not eating any more than I used to — is half-right and half-wrong. The gold-standard intake studies show systematic underreporting by about half, unconsciously. The audit is necessary even when the memory is honest, and the audit is not an accusation; it is the cheapest way to see the part of the picture you cannot see directly Lichtman et al. 1992.
Where the workup goes wrong
- The single-axis fixation. Patient and clinician chase the thyroid because thyroid is the famous answer. The result comes back normal. They stop. The sleep apnea, the medication burden, the steroid puffer were always the bigger drivers.
- Missing the prescription as the cause. A psychiatrist starts olanzapine; the primary care chases thyroid panels for two years while the antipsychotic adds ten kilograms. The medication review is the highest-yield single step in many real workups Domecq et al. 2015.
- The brush-off. "You're getting older / had a baby / it's stress" without an actual workup is the most common failure of all. Walk in with the list. Do not leave without the bloods drawn.
- Over-investigation. The inverse failure: scanning for tumours and running rare-endocrinopathy panels on a six-hour-sleeping shift worker with no other features. The pretest probability has to drive the test order; insist on the basic bloods first.
- The treat-the-thyroid trap. A mildly elevated thyroid blood test with no symptoms gets levothyroxine, the weight does not move, and the patient concludes "thyroid treatment did not work" — when the milder grey-zone version was never going to be the explanation, and the cause was somewhere else on the list Garber et al. 2012.
What identifying the cause buys you
The pattern across causes is the same: energy, sleep, mood, and cardiometabolic risk return before the weight does.
Within weeks. The cold hands warm up. The 3pm wall is gone — the meeting you used to dread goes differently. The morning fog lifts a week before the scale registers anything. The partner stops sleeping in the other room. The internal monologue that ran I must be lying to myself about what I eat quiets; the question becomes tractable Karmisholt et al. 2011.
Within months. The cause has a name and a defined treatment: levothyroxine, the mask and machine for sleep apnea, a swap from olanzapine to a weight-neutral antipsychotic, a steroid taper, the perimenopause reframe with resistance training and protein. The trousers fit again. The face that looks back from a Sunday-morning photo is the one you recognise. Stairs without thinking about it. The marriage is — somehow — better Peppard et al. 2013.
Within a year or two. The cardiometabolic markers that were sliding the wrong way stop sliding. The blood pressure that was creeping is back down. The HbA1c that was sitting at the diabetic edge has reversed. You have agency back: the shape of your body is something you are participating in again, not something happening to you Marcus et al. 2008Luppino et al. 2010.
Honest about latency: not every reader walks out of the workup with a clean diagnosis. A real fraction of "unexplained" gain on careful audit turns out to be drift the person could not see — and finding that out is still a win, because the question becomes tractable. The intervention that fits the cause becomes obvious. The next year stops being a fog Lichtman et al. 1992.
Once a cause is named, the downstream entries take over: how to dose and titrate thyroid hormone replacement, how to live with a mask and machine at night for sleep apnea, how to preserve muscle and rein in abdominal accumulation through perimenopause, when to consider the GLP-1 class of weight-loss medicines, and the body of work on dietary patterns and exercise volume for body composition. This entry is the diagnostic gate that routes you into the right one.
Substance and claimed effects
Unexplained weight gain — a sustained rise in body weight, conventionally defined as at least 5% of baseline over 6–12 months, that the person cannot attribute to a change in diet or activity. As a clinical signal, the differential is wide but largely discoverable. The standard endocrine workup considers: primary hypothyroidism Garber et al. 2012, obstructive sleep apnea Peppard et al. 2013, polycystic ovary syndrome Teede et al. 2018, Cushing's syndrome Nieman et al. 2008, hypogonadism, menopause / perimenopause Greendale et al. 2019, depression (bidirectional with adiposity) Luppino et al. 2010, medication side-effects across at least nine drug classes Domecq et al. 2015, fluid retention from heart failure / nephrotic syndrome / venous-lymphatic disease, sleep curtailment Cappuccio et al. 2008, and metabolic adaptation after prior dieting Fothergill et al. 2016. Claims attached to the entry: the symptom is a meaningful signal worth investigating rather than dismissing; a focused workup (TSH, a medication review, a sleep-symptom check, a fluid-status check, a 7-day food and activity audit, sex-and-age-specific axes) yields an addressable cause in a meaningful minority of patients; treating an identified cause typically restores some of the gained weight along with energy, mood, sleep quality, and longer-term cardiometabolic risk. Consequences in scope: short-term wellbeing, energy, focus, sleep, mood, cumulative aesthetic trajectory, and longevity — each tracks the identifiable causes the symptom flags.
Evidence by addressing question
Mechanism — how each cause produces the gain
Hypothyroidism. Low circulating thyroxine reduces resting energy expenditure by roughly 15–40% at clinically overt disease and produces myxoedematous fluid retention; on treatment, the early weight loss is mostly water and only a small fraction is fat, with median total loss of ~4 kg at one year Karmisholt et al. 2011. Subclinical hypothyroidism produces a much smaller effect; the contribution to gain in that range is real but modest Garber et al. 2012.
Obstructive sleep apnea. Bidirectional with adiposity. Repeated nocturnal hypoxia and sympathetic surges raise cortisol and insulin resistance; the chronic sleep fragmentation independently elevates ghrelin and depresses leptin, producing measurable next-day hyperphagia Spiegel et al. 2004. Population prevalence of moderate-or-worse OSA in adults aged 30–70 is approximately 13% of men and 6% of women Peppard et al. 2013.
Sleep curtailment without OSA. Habitual short sleep (≤6 h) increases the risk of incident obesity in adults by a pooled odds ratio of about 1.55 in a meta-analysis of 30 cohorts Cappuccio et al. 2008. Mechanism: appetite-hormone shift (ghrelin ↑, leptin ↓), reward-circuit sensitisation to palatable food, and more waking hours available for eating Spiegel et al. 2004.
Cushing's syndrome. Endogenous or exogenous glucocorticoid excess drives central adiposity, supraclavicular fat pads, dorsocervical pad, proximal muscle wasting, hypertension, hyperglycaemia, and skin atrophy. Rare (annual incidence ~0.7–2.4 per million) but consequential and often missed for years; the Endocrine Society 2008 guideline recommends initial screening with one of late-night salivary cortisol, 24-hour urinary free cortisol, or 1-mg overnight dexamethasone suppression Nieman et al. 2008. Exogenous Cushing's from prescribed glucocorticoids is far more common than endogenous and is missed because the cause is the patient's own prescription.
Polycystic ovary syndrome. Insulin resistance is the central driver: hyperinsulinaemia promotes androgen production and abdominal adiposity, producing weight gain that resists conventional approaches. Affects 8–13% of reproductive-age women under the Rotterdam criteria Teede et al. 2018.
Menopause transition. The Study of Women's Health Across the Nation (SWAN) followed body composition longitudinally and found that total weight rises only modestly through the transition while fat mass and visceral fat rise sharply and lean mass falls — a body-composition shift driven by oestradiol decline, not a net energy-balance change Greendale et al. 2019. The conventional reading "menopause makes you gain weight" conflates this redistribution with absolute gain Davis et al. 2012.
Medications. A 2015 systematic review and meta-analysis catalogued the obesogenic burden across nine drug classes; the worst offenders in routine use include several second-generation antipsychotics (olanzapine, clozapine), several antidepressants (mirtazapine, paroxetine, amitriptyline), corticosteroids, insulin and sulfonylureas, beta-blockers, gabapentin / pregabalin, valproate, and lithium Domecq et al. 2015. Specific magnitudes from primary syntheses: olanzapine averages ~4.2 kg at 10 weeks Allison et al. 1999; mirtazapine and paroxetine produce clinically meaningful gain at one year while fluoxetine and sertraline are weight-neutral or modestly weight-losing acutely Serretti et al. 2010; non-vasodilating beta-blockers (atenolol, metoprolol) yield about +1.2 kg over years and reduce resting metabolic rate by lowering sympathetic tone Sharma et al. 2001; valproate, gabapentin, pregabalin, and carbamazepine carry meaningful obesogenic effect while lamotrigine, topiramate, and zonisamide are weight-neutral or weight-losing McKinney et al. 2011; insulin therapy in type 2 diabetes produces predictable gain through eliminated glycosuria, increased hunger, and direct anabolic action Patel et al. 2014.
Depression. Bidirectional. A meta-analysis of 15 longitudinal cohorts found baseline depression raised the risk of incident obesity (OR 1.58) and baseline obesity raised the risk of incident depression (OR 1.55), with no clear primacy Luppino et al. 2010. Mechanisms include reduced activity, altered HPA axis, sleep disturbance, and the obesogenic class of antidepressants prescribed to treat it Serretti et al. 2010.
Fluid retention. Rapid gain — kilograms over days to a few weeks — is fluid until proven otherwise. Heart failure, nephrotic syndrome, cirrhotic ascites, and primary lymphatic / venous insufficiency are the four families. Discriminating findings: dependent oedema with diurnal variation, dyspnoea on exertion or orthopnoea, ascites, proteinuria on dipstick.
Metabolic adaptation. After significant prior weight loss, resting energy expenditure stays depressed for years below what body composition predicts; the Biggest Loser follow-up cohort showed an average resting metabolic rate ~500 kcal/day below predicted six years out, with hormonal changes (leptin ↓, ghrelin ↑) that persist Fothergill et al. 2016Sumithran et al. 2011. The relevant version of "unexplained" here is the person who lost weight a year ago, regained it without obvious behavioural change, and concludes their willpower failed; the physiology was actively opposing them.
Caloric drift the patient cannot see. Doubly-labelled-water comparisons show obese subjects underreport energy intake by an average of ~47% and overreport activity by ~51% versus measured truth — not consciously, but because portion estimation, snack frequency, and liquid calories are systematically misremembered Lichtman et al. 1992. Honest energy-balance math holds: a sustained surplus of ~22 kcal/day predicts ~1 kg/year for an average adult on the validated dynamic model Hall et al. 2011. Many "unexplained" gains are explainable on careful audit; the entry's framing is that the audit is part of the workup, not a substitute for one.
Evidence — what the workup actually yields
No single study captures "diagnostic yield of workup for unexplained weight gain" because the entity is heterogeneous and rarely studied as one condition. Yield by individual axis:
- Hypothyroidism: population prevalence of clinical hypothyroidism is ~0.3%; subclinical hypothyroidism ~4.3% in NHANES III, with female and older skew Hollowell et al. 2002. Among adults presenting with new weight gain plus any non-specific symptom (fatigue, cold intolerance, constipation), the pretest probability is elevated several-fold, which is why a TSH is the cheapest single test of the workup and is universally recommended by primary-care references Garber et al. 2012.
- OSA: 13% of men and 6% of women aged 30–70 meet criteria for moderate-or-worse disease, but ~80% of cases are undiagnosed; obesity dramatically raises pretest probability and screening with a validated questionnaire (STOP-BANG, Berlin) plus home sleep study is standard Peppard et al. 2013.
- Medication review: about half of community-dwelling US adults take at least one prescription drug; given the breadth of the obesogenic list Domecq et al. 2015, an honest medication audit by class is the highest-yield single step after the patient's own intake-and-activity audit.
- Cushing's: rare, but the workup's three first-line tests (overnight 1-mg dexamethasone, 24-hour urinary free cortisol, late-night salivary cortisol) have good test characteristics and the consequences of missing it are severe Nieman et al. 2008.
- PCOS in reproductive-age women: 8–13% by Rotterdam criteria; the diagnostic workup (history, androgens, pelvic ultrasound) is well-defined Teede et al. 2018.
Treatment response across causes: hypothyroidism replacement returns a modest fraction of the gained weight (median ~4 kg of which most is water) Karmisholt et al. 2011; treating OSA with CPAP improves daytime energy and metabolic markers but does not reliably produce weight loss without concurrent behavioural change; switching from olanzapine to a weight-neutral antipsychotic recovers a substantial share of the gain over months; tapering exogenous steroids reverses Cushingoid features over months. The general pattern: identifying and treating the cause restores energy, sleep, mood, and longevity risk well before it restores all the weight.
Protocol — how to investigate
The structure of a competent self-and-clinician workup, in order:
- Confirm it is gain, not fluctuation. Weigh on a fixed schedule (morning, post-void, fasted, same scale) for two weeks; only a sustained rise above normal day-to-day noise (1–2 kg) is the signal.
- Pace check. Gain over days–weeks is fluid until proven otherwise; gain over months–years is composition. The two need different workups.
- Medication audit. List every prescription, over-the-counter, hormonal, and PRN drug. Cross-check against the obesogenic-class list Domecq et al. 2015. Time the onset of gain against any new start or dose change.
- Sleep audit. Hours per night, snoring, witnessed apnoeas, daytime sleepiness, partner reports. A high pretest score on STOP-BANG warrants a sleep study Peppard et al. 2013.
- Intake-and-activity audit. One honest week of weighed food, including liquids and weekend pattern, against tracked activity. Most "no change" turns out to be drift — a new commute, a new coffee order, a missing gym block — that the audit surfaces.
- Bloods. Primary care first-line: TSH (free T4 reflex), fasting glucose and HbA1c, lipid panel, basic metabolic panel, complete blood count. In women with cycle changes, hirsutism, or acne, add testosterone, sex-hormone binding globulin, prolactin. If features suggest Cushing's, screen per the Endocrine Society guideline Nieman et al. 2008.
- Sex- and life-stage specifics. Perimenopausal women: track cycle pattern and consider the body-composition reading (visceral fat rising, total mass roughly stable) Greendale et al. 2019. Men with low libido, erectile dysfunction, fatigue: morning total testosterone × 2 mornings. Diabetic patients on insulin or sulfonylureas: discuss class change with the prescriber Patel et al. 2014.
- Red-flag escalation. Rapid gain with dyspnoea / orthopnoea / pedal oedema → urgent evaluation for heart failure. Gain plus easy bruising, proximal weakness, purple striae → urgent Cushing's workup. Gain plus severe headache, visual changes, galactorrhoea → pituitary workup.
Misconceptions
- "My metabolism crashed at 40." Resting metabolic rate stays flat from roughly age 20 to 60, then declines at about 0.7%/year on doubly-labelled-water data — the dramatic mid-life slowdown most adults believe in is not visible in the measurements. Reduced activity and lean-mass loss are the usual drivers of the felt change, not a metabolic cliff.
- "It must be my thyroid." Clinical hypothyroidism is real and worth ruling out, but it explains a minority of unexplained gain even in symptomatic populations and accounts for only a few kilograms when it does Karmisholt et al. 2011. A normal TSH rules it out as the dominant cause.
- "Menopause makes you fat." The transition redistributes fat to the abdomen and reduces lean mass without large total-weight change; conflating redistribution with absolute gain misframes the problem Greendale et al. 2019Davis et al. 2012.
- "My weight loss failed because I lack discipline." Post-loss metabolic adaptation depresses resting energy expenditure by several hundred kilocalories for years, with persistent hormonal hunger drive Fothergill et al. 2016Sumithran et al. 2011. The physiology is opposing the maintenance, not failing it.
- "I'm not eating any more than I used to." Doubly-labelled-water studies show systematic underreporting by ~47% in obese subjects, mostly unconscious Lichtman et al. 1992. The audit is necessary even when memory is honest.
- "The scale jumped two kilos overnight, so something is wrong." Daily fluctuations of 1–2 kg from hydration, glycogen, sodium, and stool are normal. The signal is the trend over weeks, not the spike.
Failure modes — where the workup goes wrong in practice
- Single-axis tunnel vision. Patient and clinician fixate on TSH because thyroid is the famous answer, get a normal result, and stop — missing OSA, medication burden, or steroid use that was always the bigger driver.
- Missing the prescription as the cause. A psychiatrist starts olanzapine; the patient's primary care chases thyroid panels for two years while the antipsychotic adds 10 kg. The medication review is the highest-yield single step in many real workups Domecq et al. 2015.
- The brush-off. "You're getting older / had a baby / it's stress" without an actual workup is the most common failure. The differential is too consequential to skip.
- Over-investigation. Inverse failure: scanning for tumours and running rare-endocrinopathy panels on a six-hour-sleeping shift worker with no other features. The pretest probability has to drive the test order.
- Treat-the-thyroid trap. A mildly elevated TSH in the 4–10 range with no symptoms gets levothyroxine, the weight does not move, and the patient concludes thyroid treatment "did not work" when subclinical disease was never going to be the explanation Garber et al. 2012.
Stakes — what unaddressed unexplained gain costs
Each cause has its own consequence stream, and the entry's stakes section will project them as felt experience: untreated hypothyroidism progresses to overt symptoms (cognitive slowing, depression, cold intolerance, eventually myxoedematous coma in extreme cases). Untreated moderate-or-severe OSA raises hazard ratios for incident hypertension, atrial fibrillation, stroke, and all-cause mortality Marcus et al. 2008. Steroid-driven Cushing's, if missed, produces osteoporosis, diabetes, and immune compromise within years. Antidepressant-driven gain destabilises the cardiometabolic baseline of patients already at higher cardiovascular risk. The downstream of metabolic-adaptation re-gain is the well-mapped weight-cycling literature on cardiovascular and psychological harm. The visceral-fat redistribution at menopause raises diabetes incidence (RR ~1.4) and contributes to the post-50 cardiovascular jump in women Schwingshackl et al. 2017.
Payoff — what an identified cause returns
The pattern across causes: identifying and treating the cause restores energy, sleep, mood, and cardiometabolic risk profile before it restores all the weight. Thyroid replacement: cold intolerance, fatigue, and brain fog resolve over weeks; weight loss is modest. CPAP for moderate OSA: daytime sleepiness, blood pressure, and mood improve within months; weight stays where it was without a parallel intake change. Switching from olanzapine to a weight-neutral antipsychotic: appetite normalises within weeks; about half the gain reverses over months. Tapering exogenous steroids when feasible: Cushingoid features reverse over months. Identifying perimenopause as the framing reorients the strategy from "lose weight" to "preserve lean mass and prevent visceral accumulation" — different exercise and protein priorities, different outcomes. The reframing alone — "the body is not betraying you; a discoverable cause is" — moves the patient from self-blame back to agency Luppino et al. 2010.
History
The term "unexplained weight gain" is a primary-care framing rather than a research entity; the modern workup synthesises endocrine, sleep-medicine, cardiometabolic, and psychiatric pathways that were historically separate. Twin and adoption studies established adult body weight as substantially heritable (~70% in twin data) Stunkard et al. 1986, reframing population variation as the baseline against which acquired causes are read.
Out-of-scope candidates
Treatment of any specific identified cause — hypothyroidism replacement protocols, CPAP titration, GLP-1 and bariatric pathways for refractory obesity, sex-hormone replacement decisions in menopause — belongs in dedicated entries. The general weight-management literature on dietary patterns and exercise is also out of scope. This entry is the diagnostic gate that routes the reader into the right downstream entry.
The credibility range
Optimist case. Most adults reporting unexplained weight gain have one or more discoverable causes that a competent workup will surface: a missed medication effect, undiagnosed OSA, frank or subclinical thyroid dysfunction, depression with adipogenic treatment, perimenopause's body-composition shift, post-diet metabolic adaptation, or a polypharmacy stack whose obesogenic burden no one has summed. Each cause has guideline-grade tests, a defined treatment, and a measured reversibility curve. The entry's value is high: it converts a vague, blame-loaded experience into a structured medical question with addressable answers Garber et al. 2012Peppard et al. 2013Domecq et al. 2015.
Skeptic case. On rigorous audit, the majority of "unexplained" weight gain turns out to be ordinary positive energy balance the patient could not see — caloric drift of a few hundred kilocalories per day from portion creep, liquid calories, restaurant frequency, or activity slippage Lichtman et al. 1992. Diagnostic yield for any single endocrinopathy in a population of overweight adults is low: clinical hypothyroidism is rare and produces only modest reversible weight when treated Karmisholt et al. 2011; Cushing's is rare; PCOS is sex-restricted; medication-induced gain is often known but unfixable. The framing "medical workup will explain it" risks legitimising avoidance of the harder behavioural change that explains most cases, and risks over-testing on low pretest probabilities.
Author's call. Both cases are partly right and the synthesis is the actual recommendation: the workup is high-value and under-done in primary care, and the intake-and-activity audit is part of the workup rather than a separate question to be deferred until medical causes are excluded. A reader who runs the medication review, the sleep audit, the bloods, and the honest one-week food log in parallel is being maximally responsible to both possibilities at once. The evidence rating is solid (multiple guideline-grade workup pathways, strong literature on each cause); the controversy rating is low — the field broadly agrees on what to investigate; disagreements are about emphasis (how aggressively to pursue subclinical thyroid disease, how to weight community evidence on insulin resistance).
Stakeholder and incentive map
- Endocrinology and sleep medicine. Aligned with thorough workup; their referrals depend on it.
- Primary care. Under time pressure; the brush-off failure mode is structural, not malicious. The reader's pre-formed workup ask shifts the visit.
- Weight-loss / pharma industry (GLP-1, bariatric). Indirectly incentivised to treat the symptom — weight — rather than chase the cause, since GLP-1 efficacy is similar across causes. The catalogue's role is to make sure the cause is named first.
- Wellness / metabolism-influencer ecosystem. Sells the "stuck metabolism" story; the doubly-labelled-water literature largely refutes the age-related metabolic-cliff claim. The skeptic reading of this content stream is appropriate.
- Psychiatric prescribers. Counter-incentive: switching off an obesogenic antipsychotic / antidepressant is clinically risky if it destabilises the underlying condition. The trade-off is real and the entry will name it.
Population variability
- Sex. Hypothyroidism is roughly 5–8× more common in women Hollowell et al. 2002; PCOS is women-only; menopause is women-only; OSA is more prevalent in men at any given BMI Peppard et al. 2013; hypogonadism work-up is sex-specific.
- Age. Subclinical hypothyroidism prevalence rises with age Hollowell et al. 2002. Perimenopausal redistribution begins in the early-to-mid 40s. Sarcopenic loss of lean mass accelerates after 60 and reshapes "weight" away from useful tissue.
- Polypharmacy. Older adults and adults on psychiatric care carry larger obesogenic medication burdens and benefit most from the medication review Domecq et al. 2015.
- Prior dieters. Anyone with a significant intentional weight loss in the prior 1–10 years carries persistent metabolic and hormonal adaptation and should not be told their re-gain is undisciplined Fothergill et al. 2016Sumithran et al. 2011.
- Shift workers and short sleepers. The sleep axis is independently obesogenic at population scale Cappuccio et al. 2008 and is over-represented in this presentation.
- Heritability baseline. Adult BMI is highly heritable; the question is what changed against an individual's set-point, not whether someone is heavier than average Stunkard et al. 1986.
Knowledge gaps
The diagnostic yield of a standardised "unexplained weight gain" workup has not been studied as a single protocol; yield estimates are pieced together from individual axes. The relative contribution of microbiome shifts to medication-induced and post-dieting weight regain is unsettled. Mechanistic detail on how GLP-1 agonists interact with each underlying cause (whether hypothyroid, OSA, or post-dieting patients respond differently) is still emerging; current trials report aggregate efficacy, not by-cause stratification. The reversibility curve after switching off an obesogenic medication is reasonably mapped for olanzapine but sparse for many other agents. Whether subclinical hypothyroidism (TSH 4–10, normal free T4) contributes meaningfully to weight is contested; current guidelines lean toward treatment only with symptoms or other indications Garber et al. 2012. Evidence that would shift the author's call: a large prospective study of standardised workup yield across an unselected unexplained-gain population would meaningfully tighten the estimates above; a randomised trial of medication audit vs. usual care in obesogenic-poly-pharmacy patients would settle the highest-yield-step question.
Scope and brief. The brief named the substance and its links to thyroid, hormonal shifts, medications, sleep, fluid retention, and metabolism — all of which the article covers end-to-end. Coverage matches the brief; no silent narrowing.
Hardest call: action type. Considered know (awareness) and respond (protocol for when a symptom presents). Landed on respond because the typical reader arrives already noticing the gain — the entry's value is the structured workup, not the awareness alone. know would have softened the protocol section into background reading; respond keeps it operational.
Category call. Sits in medical / Healthcare rather than food or exercise because the substance is a diagnostic gate, not a diet or training intervention. The food and exercise downstream literature is explicitly out of scope.
Rating difficulties.
energyat 4 vs 3: every cause on the differential drags energy and most reverse it within weeks of treatment. Rounded up because fatigue is the single most consistent felt symptom across the entire differential, and the lift is the entry's clearest payoff for the typical reader.health_short_term,mood,sleepall at 3: each clearly tracks the identified cause but none dominates. A 4 on any one would overstate the universality.beauty_cumulativeat 2: real but indirect — it flows from whichever cause is identified and how completely the weight gain reverses. Resisted the impulse to score 0 because the long-run trajectory genuinely bends for readers whose cause is treatable.longevityat 3: each major cause (untreated OSA, hypothyroidism, Cushing's, antipsychotic-driven cardiometabolic load, midlife visceral accumulation) carries documented mortality or major-disease implications. 3 reflects the population-weighted average across the differential, not the worst-case single cause.applicabilityat 4: the awareness audience is wider than the current-symptom prevalence per the meta spec's avoidance / recognition rule — most adults will encounter unexplained weight gain in themselves or a close other at some life stage. Stopped short of 5 because the entry's framing assumes adulthood and a baseline medical-system relationship.
Separate-entry candidates surfaced during the write.
- Hypothyroidism diagnosis and treatment — TSH interpretation, levothyroxine titration, free T4 / T3 controversies, subclinical-disease decisions. Distinct workup and management.
- Obstructive sleep apnea — STOP-BANG, home vs in-lab sleep studies, CPAP adherence and alternatives, positional therapy. The single highest-yield finding; deserves its own deep entry.
- Antipsychotic-induced weight gain and the swap decision — psychiatric stability vs metabolic load, metformin adjunct, switch protocols.
- Perimenopause and body composition — protein, resistance training, hormone therapy decisions specific to the visceral-fat redistribution.
- Metabolic adaptation after weight loss — the maintenance physiology and what the GLP-1 era is doing to it.
- GLP-1 receptor agonists for obesity — semaglutide / tirzepatide efficacy, side effects, discontinuation regain.
- Cushing's syndrome recognition — the rare endogenous case and the common iatrogenic (steroid-driven) case.
- Polycystic ovary syndrome — Rotterdam criteria, insulin resistance management, fertility considerations.
Future-link candidates. Once the downstream entries exist, this entry should cross-link to: sleep apnea, thyroid testing, perimenopause, GLP-1 agonists, weight cycling / metabolic adaptation, screening for diabetes (HbA1c).
Things deliberately left out.
- Bariatric surgery. Out of scope at the diagnostic gate; belongs in a downstream refractory-obesity entry.
- Microbiome contributions. Mechanism is plausible and emerging but the evidence is not yet at workup-actionable level; flagged in the dossier as a knowledge gap.
- Specific dietary patterns (Mediterranean, low-carb, time-restricted eating). The entry is the gate; choosing the eating pattern is a separate question that the gate routes the reader toward.
- Pediatric and adolescent presentations. Different differential (growth, pubertal timing, school-life factors); needs its own entry.
- Pregnancy-related and immediate-postpartum gain. Physiological, expected, and follows a different timeline; out of scope.
Dream-narrative tier. Overall score landed around 50 by the weighted formula, comfortably above the 40 threshold; dream narrative is obligatory. Lever: relief / debunking — the entry's hook is freeing the reader from the willpower-failure narrative, not selling an aspirational transformation. The dek and tagline carry that relief lever directly.
Voice call. Resisted the impulse to soften the stakes section toward reassurance. Each cause's silent compounding is real and the reader needs to feel it; the empathy in §1 of the voice rules is scoped to constraint the reader cannot change (a newborn, two jobs), not to the workup, which every reader can pursue.
Unexplained Weight Gain
One competent workup cycle: a medication audit, one honest week of food and sleep tracking, a primary-care visit with bloods. Done once and revisited if treatment changes. Trivial relative to ongoing behavioural interventions.
Fatigue is the dominant accompanying symptom of nearly every cause on the differential — hypothyroidism, OSA, depression, persistent post-dieting metabolic adaptation, sedating medications. Identifying the cause typically lifts the felt energy floor noticeably within weeks (Karmisholt 2011; Spiegel 2004; Luppino 2010).
Primary-care visit, TSH plus basic metabolic panel plus HbA1c, and a sleep study if indicated — most covered under standard insurance; out-of-pocket typically in the $50–500/year range. Higher if uninsured or if endocrinology workup escalates.
Each individual axis on the differential is guideline-grade (Garber 2012 for hypothyroidism; Peppard 2013 and AASM for OSA; Nieman 2008 for Cushing's; Teede 2018 for PCOS; Domecq 2015 systematic review for medication effects). The composite 'unexplained-weight-gain workup' is not studied as a single protocol; the strength is in the individual pathways.
Identifying and treating the cause delivers clear functional improvement in weeks. Levothyroxine resolves cold intolerance and constipation; CPAP cuts daytime sleepiness and blood pressure; switching off mirtazapine quiets the hunger drive. The wellness improvement is robust across causes (Karmisholt 2011; Peppard 2013).
Several of the differential's most common causes — untreated moderate-or-severe OSA, persistent hypothyroidism, exogenous-steroid Cushing's, antipsychotic-driven cardiometabolic load — carry meaningful mortality and disease-incidence implications. Catching them through the workup converts to disease-prevention years (Peppard 2013; Marcus 2008).
OSA is one of the highest-yield findings on the workup; treatment with CPAP transforms sleep quality and architecture. The sleep-curtailment axis independently drives the symptom (Cappuccio 2008) and is identified by the same audit (Peppard 2013).
Bidirectional loop: depression both produces unexplained gain and is produced by it (OR ~1.55 each direction in Luppino 2010). Hypothyroidism mimics depression. Identifying the cause stabilises mood directly — by treating the endocrine driver and by replacing self-blame with a discoverable explanation (Luppino 2010; Garber 2012).
When the workup surfaces a treatable cause — thyroid replacement, CPAP for OSA, switching off an obesogenic antipsychotic — the long-term body-composition trajectory bends back toward baseline over months to years. Modest but real cumulative aesthetic dividend tied to whichever cause is identified (Karmisholt 2011; Domecq 2015).
Hypothyroidism produces measurable cognitive slowing; OSA fragments attention through nocturnal arousal. Treating either recovers focus, but the entry's central payoff isn't cognitive — it's energy and metabolic. Real but small (Garber 2012; Peppard 2013).