Differences between HES and NHS admitted patient waiting times data

There have always been differences between the official, admitted patient waiting times figures collected and published by the Department of Health (DH) and the waiting times taken from HES. These differences are due to the different ways in which waiting times are calculated and how suspensions and deferrals are dealt with.

HES waiting times are worked out as the maximum waiting time at a given provider, such as a hospital. This time is from the date at which the decision is taken to put the patient on a waiting list for admission, until the point at which they are admitted and the treatment takes place. The HES waiting times include days when the patient isn't available for treatment, due to illness or holidays, for example. Such days are called suspensions or deferrals from the waiting list.

Before December 2003, all DH official waiting times were worked out as the date that a patient was put on a trust's waiting list to the date at which they were admitted to that trust for treatment. Unlike the HES waiting times, periods of suspensions and deferral were not included in the overall waiting time. However, there were problems with the method for working out official waiting times, including:

• Definitions for calculating the time patients were suspended or deferred were unclear, and led to inconsistencies and local interpretations.
• There were suggestions that some trusts were using 'phantom' suspensions to keep patients off waiting lists so that Government targets could be met.
• The system gave no way of reflecting the overall waiting time if a patient moved from a waiting list at one trust to a list at another.

After considerable discussion, a Data Set Change Notice (DSCN 37/2003) was issued to the NHS in December 2003 for immediate action. The changes set out in the DSCN made sure that the DH official waiting times more closely reflect the patients' perceptions of how long they have waited. The DSCN and background documents give the official guidance on this¹.

How long has a patient waited?

The new DH official waiting times are worked out from the date at which the patient is placed on a waiting list at any provider until the date at which the patient is admitted for treatment (regardless of the provider giving treatment). For example, with their consent, the patient may be moved from the waiting list at one provider to that of another provider, who may be able to treat them more quickly. This can often be a result of the Patient Choice initiative - the way HES waiting times are worked out has not changed (see previous section).

The change to the official waiting times doesn't mean that they are calculated as the difference between the date when the patient is added to the waiting list (decision to admit date) and the procedure/intervention date. However, if the treatment does not take place, for example if the patient is sent home with a cold, then the waiting time continues to be counted until the date of the admission that does result in the treatment taking place.

When are waiting times shorter than the actual waiting time?

In some circumstances, the waiting times can be shorter than the time a patient has actually waited. For example:

• Patients may decide to remove themselves from a waiting list at one provider and are subsequently added to a waiting list of a new provider (for the same treatment). In such cases, the waiting time is from the decision to admit date at the new provider.
• Patients can be suspended from a waiting list for short periods of time while they are unavailable for admission for social or medical reasons, although the latter should be kept to a minimum.
• Patients may refuse at least two admission dates that they are offered. In this case they are considered to have self-deferred and the waiting time is then counted from the earliest date offered.
• Patients may either cancel appointments or not turn up on an accepted admission date they had previously accepted. In this case the waiting time is reset to zero and the clock restarts from the date on which the patient failed to attend.

The total number of days for periods in which the patient was suspended or was self-deferred can be subtracted from the total waiting time.

Waiting times and the HES system

HES data contain four pieces of information relating to waiting times:

• Date of admission (admidate)
• Admission method (admimeth)
• Decision to admit date (elecdate)
• Duration of elective wait (elecdur).

HES does not contain 'Decision To Admit Date At Any Provider' data (also referred to as 'Original Decision To Admit Date'), as this isn't currently recorded by the Admitted Patient Care Commissioning Data Sets (APC CDS). Related changes to the APC CDS are unlikely to be made before the introduction of NHS care records.

Date of admission

The date of admission data show the date that the patient was admitted to hospital.

Admission method

The admission method data distinguish between elective admissions (which are further broken down into waiting list, booked and planned admissions), emergency admissions, maternity admissions and other admission methods.

Planned admissions are used where a patient is waiting for treatment for clinical reasons, rather than until resources become available. Such patients are usually excluded from waiting times calculations. There is no reason why the time waited should not be recorded however, as the time between clinical events could have significance.

Decision to admit date

The 'Decision to admit date' field contains the date that a consultant, or other member of staff, decided to admit a patient to hospital. The admission may be straight away, in the case of emergency treatment, or at a later date, which would be an 'elective admission'.

Duration of elective wait

'Duration of elective wait' has always been a field that the NHS was expected to include in the APC CDS. The old HES system however always recalculated it as the difference between the 'Decision to admit date' (elecdate) and 'Admission date' (admidate). HES did not examine or use the data submitted by the NHS.

Since the 2002-03 data year, HES has kept the NHS submitted value of 'Duration of elective wait' as the HES field 'elecdurn', which is held in the HES data warehouse within the system data class.

The NHS value should be calculated locally and should reflect the patient's perceived waiting time from the original decision to admit date at any provider until the actual admission date. Any periods of suspension or self-deferral should be combined and subtracted from the total.

However, it is quite possible that where a transfer has taken place, the 'new' provider will only have the original decision to admit date, their own decision to admit and admission date, and any suspensions or deferrals that have taken place at their trust. This means that the total suspensions and deferrals may not include any days from the original trust.

Comparing HES and NHS derived waiting times

In theory, it is possible for the HES waiting time (elecdur) to be greater than, equal to, or less than the NHS derived waiting time (elecdurn). The following examples may help to explain how.

Example 1: Greater than

Miss Rhodes is placed on the waiting list of Walford Infirmary NHS Trust. Unfortunately, on leaving the hospital she slips in the street and breaks her leg. This means she cannot be admitted and no offer of an admission date is made until her leg is healed. This accounts for 40 days of suspension.

At the end of her suspension period she is planning to go on holiday for three weeks and is unavailable for admission until she returns. Walford offers her an admission date for when she returns and suspends her from the admission list for the 21 days she is away.

Miss Rhodes has been suspended from the waiting list for a total of 61 days, which Walford deducts from her total waiting time. In this instance, the HES waiting time will be greater than the NHS derived waiting time as HES waiting time doesn't allow for suspensions.

Example 2: Equal to

Mr Price is put on a waiting list at Holby City NHS Trust. He is immediately offered an admission date, which he accepts. In this case, the NHS derived waiting time (elecdurn) will be the same as the HES waiting time (elecdur) as there has been no other healthcare provider involved and the waiting time has not been suspended, nor has Mr Price self-deferred his admission.

Example 3: Less than

Mrs Shurlock is added to a waiting list at Borchester NHS Trust. The Trust cannot treat her within a reasonable time and offers her the option of being treated at Ambridge General NHS Trust, which she accepts. She is subsequently treated with no further delays. Ambridge records the 'Decision to admit date' for the Admitted Patient Care CDS as the date they took over her case and the 'Duration of elective wait' as the time from which she was put on Borchester's waiting list.

In this example, the NHS derived waiting time (elecdurn) will be greater than the HES waiting times (elecdur), as the NHS derived waiting time will be calculated from the date at which she was placed on Borchester's waiting list.

Should NHS supplied or HES calculated waiting times be used for analysis?

There isn't a single answer to this question. When analysing HES waiting times data either the HES calculated or the NHS supplied waiting times can be used. There are some things, outlined in the following information, which should be taken into consideration though.

There is no way for HES to verify the value that the NHS has populated the field with and, as the previous examples show, there is no easy way of cross checking with the HES waiting times. If the value in the NHS derived waiting time field (elecdurn) is an integer between 0 and 8887 then it is assumed it is accurate. Also, as the DSCN was issued mid-year for 'immediate implementation' there is also no way of knowing whether the 2003-04 data is calculated to the old or new definitions for an individual episode.

Analysis of the 2002-03 annual refresh data suggests 143,000 Finished Consultant Episodes (FCEs) with an elective admission method have 'not known' values for the NHS supplied waiting time, ie outside the range 0000 to 8887. These are made up of mainly negative values and a series of probable default codes (9998, 9999, -999), but they account for only 2.5% of the NHS derived waiting times. Almost all the invalid values are the probable default codes. Some values are also probably misaligned; for example, 91 rather than 0091 as the data are not currently cleaned in any way.

Of the electively admitted episodes with valid waiting times, ie episodes where the HES calculated waiting time and the NHS supplied waiting time both fell within the range 0001 - 8887 days, almost all had the same calculated and supplied waiting time. This either suggests that there have been no suspensions or deferrals, or that trusts calculated the value in the same way that HES does. Of the remainder (less than 1%) there were marginally more where the HES calculated waiting time was greater than the NHS supplied wait.

1. Although the DSCN referred to changes in the value to be used to populate the 'duration of elective wait' (HES item elecdur) field, it did not specifically say that the APC-CDS should be amended. This matter is awaiting clarification from the Data and Information Standards team.