Treatment Administration Cost

Treatment Administration Cost: What It Costs to Deliver a Medicine in Economic Models

Treatment administration cost, usually shortened to administration cost, is the value of the resources used to give a medicine to a patient once it has been bought: nurse and pharmacy time, an infusion chair or day-case bed, consumables and the clinic that houses them. In an economic evaluation it sits beside the acquisition cost of the medicine and can decide whether a subcutaneous or oral option looks cheaper than an intravenous one. This page covers the cost of administering treatment, not the administrative overheads of insurers or hospitals (see administrative cost ratio). It explains what the cost includes, how it changes with the route of administration, the UK unit cost sources, how NICE committees have judged it, a worked example and common errors.

What sits inside an administration cost

An administration cost is built from the time and materials consumed at each dose. The NICE technology appraisal manual (PMG36) asks for all relevant costs, naming the costs of treatment, monitoring, staffing, facilities and training among them (section 4.4.10). For an infusion the main components are listed below.

• Staff time. Active nurse time for cannulation, setting up, observation and discharge, plus pharmacist and technician time where a dose is prepared on site.
• Chair or bed time. The infusion chair or day-case bed is occupied for the whole visit, including observation after the dose.
• Consumables. Lines, pumps, syringes and dressings.
• Facility and overheads. The unit, its equipment and its support staff.

Time and motion studies separate these parts by observing real sessions. In a study run for Roche alongside the PrefHer trial, De Cock and colleagues recorded patient chair time and active healthcare professional time both in the treatment room and in the drug preparation area, because staff often manage several patients at once and are not occupied for the whole infusion.

How the route of administration changes the cost

Route matters because it changes how long a patient occupies a chair and how much skilled time each dose needs. Intravenous infusion carries the most: preparation, cannulation, the infusion itself and observation. A subcutaneous injection usually takes minutes, and an oral medicine needs no chair at all, although it still involves dispensing and clinic review, and oral chemotherapy delivery has its own HRG (SB11Z).

Two manufacturer-funded time and motion studies show the size of the gap. For trastuzumab in early breast cancer, a subcutaneous single-use injection device saved a mean of 57 minutes of patient chair time and 13 minutes of active staff time per session compared with intravenous infusion. A later study of pembrolizumab in metastatic non-small-cell lung cancer, by the same lead author, found that active staff time fell from a weighted mean of 25.8 minutes intravenously to 14.0 minutes subcutaneously, and chair time from 117.2 to 59.0 minutes.

Home administration moves the cost again. When a manufacturer funds a nurse to give injections at home, the NHS may carry no administration cost for those doses, but the service may not last or reach every patient, which is how NICE treated it in TA1126 (see below).

UK unit cost sources

Two sources supply most UK administration costs. PMG36 calls Healthcare Resource Groups (HRGs) "a valuable source of information" for estimating resource use, notes that national average HRG unit costs can reduce the need for local micro-costing, and adds that HRG data may not suit cases where the new technology and the comparator fall under the same HRG (sections 4.4.9 and 4.4.10).

The National Cost Collection. NHS England's National Cost Collection publishes the average unit cost of defined services from annual provider submissions; the 2024/25 collection is the latest. Chemotherapy delivery has its own HRGs, with separate codes for a first attendance and for subsequent elements of a cycle. The table gives national average unit costs from the 2024/25 schedule, rounded to the nearest pound.

The care setting can move the cost as much as the complexity band: SB12Z costs GBP 436 as a day case but GBP 256 as an outpatient procedure. These are costs, which differ from the prices paid under the NHS Payment Scheme that the collection helps to inform.

The Unit Costs of Health and Social Care. For bottom-up costing, the 2025 manual, published in 2026, gives hospital nurse costs at 2024/25 prices of GBP 47 per working hour at Agenda for Change band 5 and GBP 58 at band 6, rising to GBP 53 and GBP 64 when qualification costs are included. The same report holds the NHS cost inflation index that PMG36 names for uprating older costs (section 4.4.12).

How NICE committees have costed administration

Recent NICE appraisals show committees examining the choice of cost code in detail. PMG36 itself gives an administration cost as an example of a direct, intrinsic consequence of a technology (section 4.4.16).

• TA1126, natalizumab, January 2026. Infusions in the appraisal model were costed with HRG codes, so staff hours were not modelled separately. The committee accepted that subcutaneous natalizumab takes less time to give but had not seen firm estimates of the reduction, and chose a 50% reduction in administration costs against intravenous natalizumab as the most appropriate analysis available. It left company-funded home administration out of the base case, judging it implausible that everyone would stay on it long term, and counted it as an uncaptured benefit.
• TA1122, amivantamab with lazertinib, January 2026. The company costed subcutaneous amivantamab with N10AF, a 45-minute specialist nursing appointment. The external assessment group (EAG) noted that N10AF is a community nursing code and unsuitable for an outpatient procedure, and the committee concluded that SB12Z should be used. For oral subsequent treatments the company used a one-off administration cost, and the committee preferred the EAG's monthly cost.
• TA1170, daratumumab combination, June 2026. The company used a code with nurse time only; the committee applied SB12Z, which covers nurse and chair time. The company noted that intravenous isatuximab, a comparator, takes 75 minutes to 4 hours against around 3 to 5 minutes for subcutaneous daratumumab, with 15 more hospital visits in the first 2 years.

Building administration cost into a model

In a cohort model the administration cost follows the number of doses actually given, so it is tied to time on treatment rather than to a health state label. Each cycle's cost is the number of doses given multiplied by a unit cost per dose. The discounted administration cost per patient is

$$A = \sum_{t=0}^{T} \frac{p_t , n_t , u}{(1+r)^{t}}$$

where $A$ is the expected discounted administration cost per patient, $t$ indexes model cycles up to the last cycle $T$, $p_t$ is the proportion of the cohort still on treatment in cycle $t$, $n_t$ is the number of administrations per patient on treatment in that cycle, $u$ is the unit cost per administration and $r$ is the discount rate per cycle. In a partitioned survival model $p_t$ usually comes from a time to treatment discontinuation curve; in a Markov model from the on-treatment states.

When no HRG fits, the unit cost can be built from its parts:

$$u = \frac{h_s}{60} \times w_s + \frac{h_c}{60} \times w_c + m$$

where $h_s$ is active staff time in minutes, $w_s$ the staff cost per hour, $h_c$ chair time in minutes, $w_c$ the cost per chair hour, and $m$ the cost of consumables per dose, plus any pharmacy preparation not included in $h_s$. Subsequent treatments need administration costs for as long as they are given, not a one-off cost, as TA1122 showed.

Worked example: switching from intravenous to subcutaneous (illustrative)

The example is hypothetical and its numbers illustrative, although the unit costs are the rounded 2024/25 SB12Z figures from the table. A monoclonal antibody is given 18 times. The intravenous form is costed at the day-case SB12Z cost of GBP 436 and the subcutaneous form at the outpatient SB12Z cost of GBP 256, assuming, as the EAG did in TA1122, that the subcutaneous dose is given as an outpatient procedure. The difference therefore reflects the care setting, not a route-specific HRG.

1. Intravenous. 18 × 436 = 7,848, so GBP 7,848 per patient.
2. Subcutaneous in hospital. 18 × 256 = 4,608, so GBP 4,608, a saving of 7,848 − 4,608 = 3,240.
3. A 50% rule instead. Halving the intravenous cost gives 436 / 2 = 218 per dose and 18 × 218 = 3,924, a saving of 7,848 − 3,924 = 3,924.
4. Home administration. If the first 6 doses are given in hospital and the rest by a company-funded nurse at no NHS cost, the cost is 6 × 256 = 1,536, a saving of 7,848 − 1,536 = 6,312.
5. Effect on the ICER. Suppose the subcutaneous product is a new medicine whose incremental cost before administration is GBP 10,000 against an intravenous comparator, with a gain of 0.4 QALYs. Costing both arms the same gives an ICER of 10,000 / 0.4 = 25,000 per QALY. With the step 2 saving the incremental cost is 10,000 − 3,240 = 6,760 and the ICER 6,760 / 0.4 = 16,900; with step 3 it is 10,000 − 3,924 = 6,076 and 6,076 / 0.4 = 15,190; with step 4 it is 10,000 − 6,312 = 3,688 and 3,688 / 0.4 = 9,220.
6. Staff time alone. Valuing the pembrolizumab study's active staff time at the band 5 nurse rate gives 25.8 / 60 × 47 = 20.21 for intravenous and 14.0 / 60 × 47 = 10.97 for subcutaneous; the intravenous figure is about 5% of the GBP 436 day-case cost.

At an illustrative threshold of GBP 20,000 per QALY, the decision turns on whether the administration difference is costed at all. Step 6 shows that the wage cost of active staff minutes is a small fraction of an HRG cost, which also carries chair time, consumables and overheads.

Boundaries with nearby costs

Administration cost is one part of drug cost, alongside acquisition cost and monitoring cost; the acquisition page covers prices, vial wastage and confidential discounts, so they are not repeated here. Monitoring tests and the treatment of infusion reactions are usually costed separately. Patient travel and time in the chair fall outside an NHS and PSS perspective but count under a societal perspective. Using an HRG average is a form of gross costing, while timing each task is micro-costing.

Common errors with administration cost

Most errors come from picking a convenient code rather than one that matches how the dose is given. The following recur in appraisals.

1. Nurse time without chair time. A nursing contact code misses chair time (TA1170).
2. A code from the wrong setting. A community nursing code used for a hospital procedure (TA1122), or a day-case cost for a dose given as an outpatient procedure.
3. The same HRG in both arms. PMG36 warns that HRG data may not suit technologies that fall under the same HRG; micro-costing may be needed (section 4.4.10).
4. Home delivery assumed for everyone. Company-funded services may not reach all patients or last for the time horizon (TA1126).
5. Time savings counted as cash. Freed chair time is capacity, a saving in opportunity cost terms only if the time is used for other patients.
6. Old costs left unadjusted. Costs from earlier years should be uprated with an appropriate index (PMG36 section 4.4.12).

Sources

• De Cock E, Pivot X, Hauser N, Verma S, Kritikou P, Millar D, Knoop A. A time and motion study of subcutaneous versus intravenous trastuzumab in patients with HER2-positive early breast cancer. Cancer Medicine. 2016;5(3):389-397. https://doi.org/10.1002/cam4.573
• De Cock E, Oskar S, Lourdudoss C, Eiras R, Pietanza MC, et al. A time and motion study comparing subcutaneous pembrolizumab versus intravenous pembrolizumab in combination with chemotherapy for the treatment of metastatic non-small cell lung cancer. Advances in Therapy. 2025;42(12):6175-6189. https://doi.org/10.1007/s12325-025-03365-7
• Jones K, Weatherly H, Barker A, Birch S, Castelli A, Chalkley M, Dargan A, Findlay D, Hinde S, Markham S, Smith D, Teo H. Unit Costs of Health and Social Care 2025 Manual. Care and Outcomes Research Centre (previously the Personal Social Services Research Unit), University of Kent, and Centre for Health Economics, University of York; 2026. Table 11.2.2. https://doi.org/10.22024/UniKent/01.02.115569
• National Institute for Health and Care Excellence. NICE technology appraisal and highly specialised technologies guidance: the manual (PMG36). London: NICE; 2022, updated March 2026. Sections 4.4.9, 4.4.10, 4.4.12 and 4.4.16.
• National Institute for Health and Care Excellence. Natalizumab (originator and biosimilar) for treating highly active relapsing-remitting multiple sclerosis after disease-modifying therapy (TA1126). London: NICE; 2026. Sections 3.21 and 3.22. https://www.nice.org.uk/guidance/ta1126
• National Institute for Health and Care Excellence. Amivantamab with lazertinib for untreated EGFR mutation-positive advanced non-small-cell lung cancer (TA1122). London: NICE; 2026. Sections 3.18 and 3.22. https://www.nice.org.uk/guidance/ta1122
• National Institute for Health and Care Excellence. Daratumumab with bortezomib, lenalidomide and dexamethasone for untreated multiple myeloma when a stem cell transplant is unsuitable (TA1170). London: NICE; 2026. Sections 3.3, 3.10 and 3.13. https://www.nice.org.uk/guidance/ta1170
• NHS England. National Cost Collection for the NHS: 2024/25 National Cost Collection data, National Schedule of NHS Costs (Admitted Patient Care and Outpatient Procedure tables). NHS England; 2025. https://www.england.nhs.uk/costing-in-the-nhs/national-cost-collection/