Who funded the trials, and how much of it came back

Each column is one modalityModalityA family of treatments that work the same way — such as radiofrequency microneedling, or lasers that leave the skin surface intact — rather than any one branded machine.; each band is the lead-sponsor classSponsor classThe registry's grouping of who ran a study — a company, the national health agency, another government body, or the catch-all bucket that holds hospitals and universities. that ClinicalTrials.govClinicalTrials.govThe United States public registry where the people running a study record it before or during the study, which is where this site's study counts come from. recorded on the study. Colours here are sponsor classes, not site categories.

Axis labels are shortened names. Terms in them: Dermal fillerDermal fillerAn injected gel that adds volume under the skin. · NeuromodulatorNeuromodulatorAn injected product that temporarily relaxes a muscle, which is the category botulinum toxin products sit in. · RFRFRadiofrequency: energy that heats tissue from the inside by passing an electrical current through it rather than by shining light on it. · BBLBBLBroadband light: one manufacturer's name for its intense pulsed light platform.

Lead-sponsor class of every registered study, per modality. Industry share is computed over studies with a KNOWN sponsor class; no study on this run was missing one.

“University, hospital or other” is the registry's catch-all class, not a third kind of funder. i

One row per modality. The bar is the middle half of that modality's actual enrolment counts — the 25th to the 75th percentile — and the dot on it is the medianMedianThe middle value once everything is lined up in order, which a few unusually large or small entries cannot drag around.. Beside each name: the median, the middle half and the number of studies both were computed from.

Actual enrolment per modality: the median, the middle half, the range, the total and how many studies reported anything at all. Planned or estimated enrolment is excluded and counted separately.

Median participants per registered trial that reported an actual enrolment; small medians mean small studies, not necessarily weak ones. i

Two views of the same rows. “Posting rate” is the share of each modality's COMPLETED registered studies carrying results posted, with the count it came from beside every name; “Completed studies” draws those same studies as counts, so a modality with two completed studies is two studies long.

Share of COMPLETED registered studies with a results section posted on ClinicalTrials.gov, with the counts the share was computed from.

Registration is not publication: a low rate means results were not posted to ClinicalTrials.gov, not that the device failed. i

Each bar is one modality's registered studies, split by how far down its ranked list of lead sponsors they sit: the largest one, the next four, then everyone else. The first two bands together are that modality's top-five share.

How few organisations the registered evidence rests on, per modality, with the single largest lead sponsor named.

Every modality's five largest lead sponsors, exactly as the registry spells them.

Sponsor names are the registry's raw lead-sponsor strings — this site does not merge them, so one company spelled two ways counts twice. i

Each bar is one modality's registered studies, split three ways by start date: on or after 2021-07-26, before it, or no start date the run could parse. The three parts are the whole.

Share of each modality's registered studies that started within the last five years, with the counts behind it.

An old evidence base is not automatically a weak one — a settled technique may simply have stopped needing new trials. i

One row per modality, four dots on it: each dot is one publication type as a share of that modality's own PubMed total, and the total is printed beside the name. The dots are separate marks, not a stacked bar, because these types overlap and do not sum to the total. Every count, including the three types the chart does not draw, is in the table under each chart.