Take-up rates of Scottish benefits - 2026

Annex C: Disability benchmark methodology

This technical annex provides more detail on the modelled eligibility benchmarks for disability benefits in this year’s take-up publication, covering the methodology and limitations of the benchmarks in more detail.

Differences between survey measures of disability and eligibility for disability benefits

The main challenge in producing take-up estimates for disability benefits is calculating a robust estimate of the eligible population^[2] (e.g. the number of people who would be awarded ADP if they applied, meaning a score of at least 8 in either Daily Living or Mobility against the ADP scoring criteria). This is unlike calculating the eligible population for other benefits like Scottish Child Payment, which can, in the main, be directly observed using administrative data.

Estimates of the disabled population

A starting point for eligibility analysis is the disabled population of Scotland, adjusted for the respective relevant age brackets for ADP and CDP. However, many disabled people will not have needs that would result in an award for ADP or CDP, meaning the survey measure of the disabled population is likely to overestimate eligibility.

For example, someone whose condition means they can stand and move more than 50 metres but no more than 200 metres and can plan and follow the route of a journey unaided, would not be eligible for the mobility component of ADP (scoring 4 points, where 8 points are required for the standard rate). However, this person is likely to meet the core definition^[3] of disability under the Equality Act 2010, which is having a physical or mental impairment that has a ‘substantial’ and ‘long-term’ negative effect on your ability to do normal daily activities.

Condition-based measures

ADP and CDP are awarded based on specific criteria that reflect individuals’ mobility and care needs. This means that data on the prevalence of health conditions does not provide an accurate guide to eligibility, because mobility and care needs can vary among people with the same health conditions.

For example, a young child with diabetes might be unable to manage their condition independently, while an older child or adult with the same condition might be able to do so without support. Needs can also arise from a combination of health conditions^[4]; for example, whether an adult with diabetes requires support to manage their condition may depend upon whether they have other health conditions that prevent them from doing so independently.

Additionally, the mobility and care needs associated with a health condition may depend upon its duration or severity. For example, the mobility and care needs of someone with a degenerative condition such as Parkinson’s disease are likely to increase over time.

Needs-based questions

In addition to survey estimates of the disabled population or the prevalence of health conditions, some surveys also contain questions on the effects of health conditions on day-to-day life. For example, the Family Resources Survey asks whether an individual’s conditions or illnesses affect vision, hearing, mobility, dexterity, learning, or understanding or concentrating, memory, mental health, stamina or breathing or fatigue, or affects them socially or behaviourally.

Questions which focus on the impact of a condition on a person’s day-to-day life are better aligned with the criteria for ADP or CDP than estimates of condition prevalence alone. However, limitations remain in using these measures to identify eligibility if the questions do not align exactly with the ADP or CDP criteria. For example, someone who reports difficulty with mobility in the Family Resources Survey^[5] might still be able to walk sufficiently far unaided that they would not receive the mobility component of ADP.

In addition, survey data reflects an individual’s own perceptions of their health and needs, which may differ from the person-centred decisions taken by Social Security Scotland. This means that even if survey questions aligned exactly with disability benefits criteria, there could be differences between self-reported data and the decisions that would be taken on applications.

Methodology

To provide a more relevant comparison than the total disabled population, we develop a new modelled eligibility benchmark. The key modelling step in this benchmark is the use of data on application success rates to act as a proxy for the eligibility rate among disabled people not in receipt of disability benefits.

Use of application success rate as a modelling tool

Applicants for disability benefits are in most cases a subset of the wider population of disabled non-recipients. Data on applications therefore provides our most relevant observable source of information about eligibility within that wider population.

Applications for disability benefits are likely to come from disabled non-recipients due to two features of disability benefits.

Firstly, someone with an award for a disability benefit does not need to reapply for that payment to continue receiving it, meaning applications are only drawn from disabled people who do not currently receive that payment^[6]. This means that when someone makes a successful application for ADP or CDP, they leave both the pool of disabled non-recipients and the pool of potential applicants.

Secondly, Social Security Scotland makes person-centred decisions on applications based on individual needs across 12 living and mobility activities and descriptors, meaning many applicants are unlikely to know the exact award decision they will receive before applying. As a result, applications are drawn from a wider pool of people who are encouraged to apply (disabled people who need help with certain activities). This means, for example, that an increase in the working-age disabled population may also increase the number of potential applicants for ADP.

The application success rate is an observable proxy for the eligibility rate among the pool of people applying for a payment, meaning it provides useful information about the eligibility rate among disabled non-recipients, particularly over time, due to the links between these groups. However, unless the pool of applicants for a benefit is fully representative of the wider population of disabled non-recipients, it will not capture the exact eligibility population.

Other adjustments

Age

It is necessary to adjust for the fact that for both ADP and CDP, the age groups eligible to apply for the benefit are not the same as the age groups eligible to receive the benefit.

The total number of ADP recipients includes pension age adults who were already in receipt of ADP before reaching state pension age. This means that when calculating the number of working-age disabled people not in receipt of ADP, we deduct only working-age ADP recipients from the working age disabled population.

Similarly, the total number of CDP recipients includes people aged 16-18 who were already in receipt of CDP at age 16. This means that when calculating the number of disabled under 16s not in receipt of CDP, we deduct only CDP recipients under 16 from the under 16 disabled population.

Other benefits

The figures in the main publication plot our modelled eligibility benchmarks across 2022-23, 2023-24, and 2024-25. During this time period, transfers of awards from reserved disability benefits to devolved disability benefits was ongoing.

This means in the time period plotted, working-age disabled people in Scotland could have been receiving ADP, PIP, or DLA^[7], while disabled children could have been receiving CDP or DLA. To allow for a consistent interpretation of eligibility benchmarks over time, our analysis focuses on the total number of people receiving either a reserved or devolved benefit. The reason for this is that an ADP eligibility benchmark would underestimate the true number of people who would receive an ADP award on applying if it excluded those currently receiving PIP who were waiting for their award to transfer^[8].

To carry out this adjustment, when we calculate the total number of benefit recipients in an age group for our eligibility benchmark, we include everyone receiving either a devolved or reserved disability benefit. Similarly, when we calculate the number of disabled non-recipients in an age group, we deduct everyone receiving either a devolved or a reserved disability benefit in that age group.

Limitations

The modelled eligibility benchmark will align with the true eligible population for a payment if:

• the success rate of applications is equal to the eligibility rate amongst disabled people not currently in receipt of that payment, and;
• the survey measure of the disabled population used is an appropriate measure of the pool of potential applicants.

The table summarises possible sources of bias in our eligibility benchmark, and indicates how this affects the benchmark relative to true eligibility, with additional detail provided below. Upwards bias means that the benchmark is likely to be larger than the true eligible population, while downwards bias means the benchmark is likely to be smaller.

Source: Barriers to access social security

Summary: Disabled people who are likely to meet the award criteria are likely to also face greater barriers in accessing the social security system.

Bias: Downwards

Source: Informed decisions to apply

Summary: In the absence of barriers, people who are eligible for a benefit are more likely to apply than people who are not.

Bias: Upwards

Source: Unsuccessful applicants

Summary: People who previously applied but were not awarded a payment are unlikely to be eligible if their needs have not changed.

Bias: Upwards

Source: Eligible people not captured by survey disability questions

Summary: People may be eligible for disability benefits without identifying as disabled, or may not be captured by survey questions.

Bias: Downwards

Source: Eligible people living outside Scotland

Summary: Some people living outside the UK can be eligible for Scottish disability benefits.

Bias: Downwards

Source: No recourse to public funds (NRPF)

Summary: Some disabled people resident in Scotland are ineligible to receive benefits due to immigration status.

Bias: Upwards

As the benchmark features both upwards and downwards sources of bias, it is not possible to say whether overall it is larger or smaller than true eligibility.

Barriers to accessing social security

Some disabled people may not make a successful application because they face barriers to accessing social security, even if their needs do meet the thresholds for awards. For example, their condition may make it difficult for them to make an application independently, or they may experience stigma accessing support for a condition.

People with greater needs are likely to be those who face greater barriers to access the social security system, because the needs that make them eligible for a benefit may also be the needs that make it more difficult to apply for a benefit. This could mean the application success rate underestimates eligibility among disabled people not currently receiving disability benefits.

Informed decisions to apply

In the absence of barriers, people are more likely to apply for benefits where they believe they are more likely to be awarded the benefit.

If people choose to apply for benefits based on their expected chance of success, this would bias application success rates upwards, compared to the true rate of unfulfilled eligibility amongst disabled people.

Unsuccessful applicants

Some disabled people who are not currently in receipt of disability benefits may have made an unsuccessful application in the past. If their circumstances have not changed, and their previous application accurately reflected their needs,

the decision on their previous application would suggest they are not currently eligible for a benefit.

Our benchmark uses the current application success rate as a proxy for the eligibility rate amongst disabled non-recipients. If many disabled non-recipients have already made unsuccessful applications, and do not reapply unless their circumstances change, then the application success rate may overestimate eligibility among disabled non-recipients.

An alternative approach would be to remove people who have previously made unsuccessful applications from the pool of potential applicants when calculating the benchmark. However, as decisions on applications are taken based on an applicant’s circumstances and the information available at the point of application, and circumstances can change over time, this would exclude some people who are eligible - for example, people whose health has deteriorated since their previous application.

Taken together, it is unclear if the application success rate overestimates or underestimates the eligibility rate amongst disabled non-recipients. Barriers to accessing social security would bias the estimate downwards, whereas people choosing to apply based on expected chances of success, or not applying because they have been unsuccessful in the past and their needs have not changed, would bias the estimate upwards.

Eligible people not captured by survey questions

Some people will be eligible for Scottish disability benefits without being captured in survey estimates of the disabled population. They may be disabled in a way that is not captured by the survey question, or they may not identify with the terminology in the question. This means survey measures of the disabled population may underestimate the pool of potential applicants.

Social Security Scotland client and applicant diversity and equalities analysis for 2025-26^[9] found that 85% of applicants for ADP are reported as having a “Physical or mental health condition or illness lasting or expected to last 12 months or more”, while 7% chose not to answer the question, and 7% reported not having a condition or illness lasting over 12 months. 34% of those who reported a condition were successful (out of those who have an outcome), compared to 27% of those who reported no condition.

Although the question is not identical to the FRS disability measure, these figures illustrate that a small but material proportion of applicants, including some successful applicants, may not be captured by a long-term-condition question of this kind.

Eligible people living outside Scotland

It is possible to receive Scottish disability benefits while living in the EEA, Switzerland, or Gibraltar, if someone has a genuine and sufficient link to Scotland. For disabled people living in the EEA or Switzerland (excluding Ireland), they must be covered by the EU Withdrawal Agreement, and so have moved before 2021 (or be joining a family member who moved before 2021).

A survey estimate of the disabled population resident in Scotland may therefore underestimate the number of disabled people who could be eligible to apply for a payment.

It is not possible to observe the number of disabled people who live outside the UK who have a substantial connection to Scotland and are covered by the EU Withdrawal Agreement. However, data on ADP recipients suggests that only a small fraction of ADP recipients are currently living outside Scotland. As of July 2025, the number of ADP recipients without a Scottish Local Authority recorded was 850 (0.18%), out of a total of 484,055^[10]. Individuals could also be in this category due to errors in postcode recording, or living in newbuild properties not yet linked to Local Authority data.

Given the small number of current recipients and the specific circumstances required to be eligible for Scottish benefits outside Scotland, the number of people with unfulfilled eligibility for Scottish disability benefits who are not resident in Scotland is likely to be small relative to the resident population.

No recourse to public funds

Working-age adults with No Recourse to Public Funds (NRPF) are ineligible to access benefits even if they meet other eligibility criteria, due to their immigration status. Survey estimates of the disabled population of Scotland may overestimate the number of disabled people who could be eligible to apply for a payment, as some disabled people in Scotland may have NRPF.

The UK Government does not systematically record data on the population with NRPF status, meaning it is difficult to understand how many disabled people in Scotland have NRPF^[11]. A UK parliament report estimates that at the end of 2024, around 3.6 million people held visas that would usually have an NRPF condition^[12]. Additionally, in June 2024, around 225,000 people were in various stages of the asylum system^[13], meaning they would have NRPF status. Estimates of irregular migration status contain greater uncertainty, with the most recent estimates (2020) suggesting around 600,000-800,000 people in the UK may have irregular migration status^[14]. Taken together, these separate estimates suggest that around 4.4 million to 4.6 million people, equivalent to around 6% to 7% of the UK population, may have had NRPF. This is an indicative calculation rather than an official estimate, and is subject to considerable uncertainty because the sources use different definitions and reference periods.

In addition, this percentage may not be an accurate guide to the percentage of disabled people with NRPF, as the prevalence of health conditions amongst people with NRPF may differ to the general population. For example, work visas require income above a certain threshold, so migrants who are able to obtain these visas are less likely to have conditions that prevent them from being economically active. This means that the proportion of the disabled population with NRPF may differ from the proportion of the general population.

Overall, it is not possible to conclude whether the Scottish disabled population as measured by the FRS overestimates or underestimates the pool of potential applicants, as some eligible people may not be captured by the FRS disability questions or may not be Scottish residents, biasing the estimate downwards, while some Scottish residents may have NRPF, biasing the estimate upwards.