What SLIPTA is

SLIPTA stands for the Stepwise Laboratory Quality Improvement Process Towards Accreditation. It is a checklist that measures how well a medical laboratory's quality management system is working, and awards a rating from no stars up to five.

It is a stepwise process on purpose. A laboratory is not expected to arrive at five stars; it is expected to be audited, see precisely where it stands, fix what the audit found, and be audited again. The rating is a measurement taken along the way, not a pass mark.

The rating is recognition, not accreditation. Accreditation is a separate, formal process carried out by an accreditation body. A star rating from this system does not confer it, does not substitute for it, and does not replace an official audit.

02

The audit journey, end to end

Plan the audit. A laboratory and a date are chosen, and the version of the checklist to be used is fixed at that moment — so re-reading an old audit years later still uses the instrument of its day, not today's.
Assemble the team and declare conflicts of interest. A conflict is a recorded field with three states, not a convention: none declared, declared and unresolved, or declared and cleared. An unresolved conflict blocks the audit from starting.
Capture, with evidence. Each question is answered Yes, Partial, No or Not applicable, with a written observation. A question that awards points needs a document attached.
Score. The score is computed as the answers are entered, by section and overall. No one types a score in.
Close with a sign-off. Only a senior reviewer may close an audit, and never the person who captured it. Closing records who signed, when, and the exact figures they signed against.
Report. The audit becomes a report: the score by section, every finding, and the evidence behind each one.
Corrective action. Each finding can become a corrective action with a root cause, an immediate correction and an action to stop it recurring — which is only complete once its effectiveness has been reviewed.
Re-audit. The laboratory is audited again, and the two audits can be compared.

Two of these are separations of duty rather than conveniences: the person who captures an audit cannot close it, and the person who performs a corrective action cannot certify that it worked. Both are enforced by the server, not only by the screen.

03

How scoring works

Every question carries a number of points. An answer is one of four values, and each does something different to the arithmetic:

What each answer does to the score
Answer	Points earned	Effect on the total it is measured against
Yes	All of the question's points	Stays in
Partial	Exactly one point, whatever the question is worth	Stays in
No	None	Stays in
Not applicable	None	Comes out — the question is removed from the total
Left blank	None	Stays in — an unfinished audit cannot flatter itself

The distinction in the last two rows is the one that matters most. Not applicable means the question could not reasonably apply to this laboratory, so it is taken out of the denominator. A blank is simply unanswered, and stays in — otherwise an audit could raise its own percentage by leaving hard questions alone.

Partial is worth a flat single point rather than a proportion. A partially met requirement worth ten points and one worth two points both earn one point.

04

Worked examples

Check the arithmetic. The full instrument is worth 367 points.

A straightforward audit

A laboratory earns 290 of the 367 points, with nothing marked not applicable.

290 ÷ 367 = 77.74%

77.74% falls in the 75.00–84.99% band, so the laboratory is at 3 stars.

When something is not applicable

The same laboratory, but questions worth 40 points in total genuinely do not apply to it. Those points leave the total. It earns 290 of the remaining 327.

290 ÷ 327 = 87.09%

The same 290 points now read as 4 stars, because they were earned against a smaller, fairer total.

Why blanks are not the same

An audit is abandoned half way. Questions worth 40 points are simply unanswered, not marked not applicable. They stay in the total.

290 ÷ 367 = 77.74%

Still 3 stars. Leaving questions blank can never raise a score.

Percentages are computed in integer arithmetic and truncated, never rounded up, so a score can never be nudged over a band boundary by a rounding rule.

05

The star levels

SLIPTA star levels by percentage of applicable points
Stars	Percentage of applicable points
5 stars	95.00% and above
4 stars	85.00% – 94.99%
3 stars	75.00% – 84.99%
2 stars	65.00% – 74.99%
1 star	55.00% – 64.99%
No stars	below 55.00%

06

The twelve sections

The twelve SLIPTA sections and their points
#	Section	Points
1	Documents and Records	22
2	Organisation and Leadership	26
3	Personnel Management	34
4	Customer Focus	24
5	Equipment Management	38
6	Assessment	24
7	Supplier and Inventory Management	27
8	Process Management	71
9	Information Management	24
10	Nonconforming Events	13
11	Continual Improvement	7
12	Facilities and Safety	57
	Total	367

Section point totals are the published header totals for the instrument. In two sections the header total differs by a few points from the sum of the individual questions beneath it; that is a known feature of the published checklist, and this system keeps the published totals rather than quietly correcting them.

07

Offline, and what happens when a signal returns

The tool does not need a connection to work. Every answer is written to the device first, and only then queued to be sent. If the connection drops mid-audit, nothing waits on it and nothing is lost.

When a connection returns, the device sends what it has and asks the server for what it is missing. Two rules make that safe: sending the same change twice can never apply it twice, and the server decides the order rather than the device's clock — so a device with the wrong time set cannot overwrite someone else's later work.

If the server refuses a change, the change is held and shown, not discarded. Anything the system cannot complete is visible on a sync screen rather than silently dropped.

How it works