Employee Assistance Program

What an employee assistance program is

An employee assistance program (برنامج مساندة الموظفين), often shortened to EAP, is a service the organisation pays for and an independent third party delivers. Employees contact it directly, without going through anyone, about matters that affect their ability to do their work, and a confidentiality line agreed in advance separates the provider from the organisation.

Three conditions have to hold together. The party paying is not the party providing. Access is direct and does not pass through a manager. And what travels from the provider back to the organisation is fixed in advance. Remove the second or the third condition and the name survives while the thing that made it work is gone.

What follows describes an administrative and contractual arrangement. What the services themselves consist of, how they are delivered, and what qualifications the people delivering them hold are matters for the provider and its contract, and nothing about them follows from the definition of the program.

The line between what reaches the organisation and what does not

The whole arrangement rests on one premise: the organisation pays and does not learn who used the service. What reaches it is a set of aggregate figures: how many people used the service, in which broad areas, and how many sessions they had on average. No name reaches it, no description of a case, and nothing of what was said.

The most delicate part of this is arithmetic rather than contract wording: the group size below which a report is not accepted. A report saying that two people used the service in a unit of six employees names nobody, and it has still pointed at a third of the unit. The same kind of report about an organisation of 600 employees, of whom 24 used the service, points at no one.

That is why the contract sets a minimum size for any group reported on, and sets it before the first report is designed, not after somebody has seen one. Leave the clause out and every breakdown in the report becomes something that can be traced back to individuals. Employees tend to notice this before the organisation does, and they stop using the service. The same logic governs subordinate appraisal, where a result is withheld until enough responses have arrived to keep any one of them from being identified.

How the rules on employees’ personal data apply to any of this is a statutory question with its own sources. The definition of the program settles none of it, and nothing here should be read as a ruling on it.

Three routes in, and they do not carry the same confidentiality

The most common confusion is to treat confidentiality as a single property of the whole program. In practice it depends on the route by which the employee arrived:

• Direct access. The employee contacts the provider on their own, and the organisation never learns that it happened. This is the default form, and employees’ willingness to use the program rests on it.
• A manager’s suggestion. The manager mentions that the service exists and leaves the decision to the person concerned. In substance this is still direct access, on one condition: the manager is never told whether the employee went. If that answer comes back to the manager, the suggestion has become monitoring under another name.
• A conditional referral. Here the employee agrees in writing that a defined amount of information may reach the organisation. The limit of that information is written down before the referral, not after it, and it is confined to three things: whether the employee attended, whether they continued, and whether the arrangement has ended. It never extends to what the sessions were about, what was said in them, or any opinion on the employee’s condition.

Confusing the third route with the first two is what brings a program down. It takes only one employee whose manager hears something the employee never agreed to release: the whole unit soon knows, and after that the service goes unused. Uptake of any arrangement of this kind follows what people expect to happen after they use it, not what its description promises.

The cost arithmetic is usually read upside down

Take an organisation with 600 employees and an annual contract at SAR 45 per employee. The cost is SAR 27,000 a year, and it is fixed: it does not move with the number of people who use the service.

• At a utilisation rate of 4%, which is 24 people, the cost per user is SAR 1,125.
• At 9%, which is 54 people, the cost per user is SAR 500.

So the program gets cheaper the more it is used, which is the reverse of how most employee benefit lines behave. It follows that reading low utilisation as good news, on the grounds that few people needed help, gets the result backwards. At that level the organisation has paid SAR 27,000 for something that reached almost nobody, and it will pay the full amount again next year.

The utilisation rates used here are assumed, chosen only to show how the cost per user moves with the count. In the sources we reviewed we found no general benchmark for a correct utilisation rate, so neither figure should be read as a target.

The more useful figure is not utilisation alone but utilisation read alongside awareness: the share of employees who know the service exists at all. Low use with low awareness is a problem of making the service known. Low use with high awareness is a problem of trust in the confidentiality line. The two call for entirely different remedies.

Use is uneven, and contracts are priced as if it were even

Go back to the 54 users and suppose that 40 of them needed a single session while 14 reached six sessions each:

• Total sessions: 40 plus 84, which is 124.
• Average per user: 124 divided by 54, which is about 2.3 sessions.
• The typical user had one session, not two or three.

An average of 2.3 describes nobody, neither the 40 nor the 14. Contracts are often priced on an average number of sessions per user, which gets the total right and misses precisely the cases the program was built for.

So read two further figures alongside the average: how many users reached the maximum available to them, and what happens to someone who reaches it. If the answer is that the arrangement simply ends at that limit, the program serves the lighter cases and stops at the heavier ones. Employees should be told that from the start rather than discover it when they need more.

Four clauses to read before the price

• Who is covered. The employee alone, or the employee and their dependants. The difference shows in the price, and it shows even more in what the service means, because much of what keeps an employee from their work happens outside it. The statutory rules on health insurance have their own sources, and the definition of the program cannot stand in for them.
• Response time. How quickly a first request is answered, and whether there is an arrangement for urgent cases. A response time that is not written down means, in practice, the longest delay the provider can get away with.
• The cap. The number of sessions available, and what happens once it is reached, as set out above.
• What comes after. Where someone is referred when their situation goes beyond what the contract provides, and who bears that cost. Leaving this clause out means the program ends at the first case it was really needed for.

These clauses are read against the expected utilisation, not against the price alone. A cheap contract with narrow terms produces a high cost per user because nobody uses it, which makes it the worse deal even when it looks like the better one in the budget. The same holds for added benefits generally: they change the conditions people work in, while what changes how people behave is narrower and harder to reach.

Where it sits among the arrangements it is mistaken for

The axis that separates it from its neighbours is what each arrangement acts on. Wellbeing at work is the employee’s condition across every aspect of the job; this program is a single service that an organisation adds. The sources of harm in the work itself are examined first and supporting services are added afterwards, and the program is one of those supporting services; it does not address the sources themselves.

A complaint is about a decision or an incident on which the organisation is asked to rule. The program rules on nothing and tells the organisation nothing, so it cannot carry a complaint. Conduct in the workplace, harassment included, has its own route, and a reporting channel of the kind described under corporate whistleblowing is built so that what it receives gets investigated. Sending such a matter to the program means, in practice, that nobody looks into it.

Change in the work sits on a different axis again. A service added on top of a workload that has not changed treats its effect on the person and leaves its cause in place. A year later that shows up in the report as rising utilisation with no improvement in anything.

What undermines it

• Announcing it once. Someone who needs it does not go looking for it at the moment of need, so awareness has to be renewed through several channels and in places where nobody can see who is reading.
• Making access depend on approval. Any administrative step before access cancels direct access, and it is usually the first thing added when someone wants to control the cost.
• Too much detail in the report. Every extra breakdown shrinks the group being reported on, with the effect described earlier.
• Crediting it with results across the organisation. Improvements in the organisation’s key performance indicators move for many reasons, and assigning them to the program alone goes beyond what the data can support.

Nothing in the definition establishes that the program reduces absence or raises any indicator by a known amount.

There is also a mistake of timing rather than design: launching the program in the middle of a restructuring or a reduction in headcount. Employees then read the service as something it is not: taking it up looks like a signal, and it is credited with an intention it does not have. At such times something with a clearer effect comes first: a plain statement of what is going to happen and when.

Before the contract is signed

The most useful test before signing is to write down the answer to one question precisely: what exactly reaches the organisation, about groups of what size, and in what form? If that answer does not set a written limit as a specific number, the clause does not exist, however often confidentiality is mentioned in the description of the service.

Then decide what the program will be measured by before it starts: the awareness rate, the utilisation rate and the cost per user. Present the three together, because any one of them on its own supports two opposite readings, and the more common of those readings turns the result upside down.

This is an explanation of the concept, not legal advice.

Related terms

• Subordinate Appraisal
• Corporate Whistleblowing
• Employee Benefits Accounting
• Exit Interview
• Employee File