Hospital at Home in Scotland evaluation: phase 1 findings report

Appendix B - Methodology

An Evaluation Advisory Group (EAG) was convened by the Scottish Government research team to support the evaluation. Membership of the EAG includes a range of stakeholders (e.g. clinical representatives). The role of the EAG includes providing advice and feedback on various matters such as the methodology proposed and final outputs.

Health and Social Care Workforce Survey

Background and methodology

The development of the workforce survey content was led by government Social Researchers from the Whole System Intelligence Analysis Division. This involved extensive engagement with a range of stakeholders, including from within the Hospital at Home Programme Board group. The survey consisted of 27 questions, 22 closed and 5 open-text (see Appendix C). The workforce survey began with three contextual questions, designed to understand the respondents role and profession. It then split into two distinct ‘routes’, one for those delivering H@H care (‘delivery’ respondents’) and one for those referring patients to H@H (‘referral’ respondents). One of the contextual questions asked respondents to identify what best reflected their “main area of work”. Those who responded “Hospital at Home team” or “Specialist team which includes delivering Hospital at Home care” followed the survey route for those delivering H@H care, whilst the remaining respondents followed the survey route for those referring patients to H@H. At the end of the survey, respondents were asked if they would be interested in being contacted with further information about participating in a focus group session. Those who expressed an interest were asked to provide their email address for information about the focus groups to allow them to be contacted.

A snowball sampling approach was used to distribute the workforce survey, which was available for online completion only. An invitation email was developed, which included the survey’s background information, the survey access link and the privacy notice, as well as a note to state that the invite was sent on behalf of the Deputy Chief Medical Officer. Given the broad and disparate nature of the target sample, a range of avenues were used to cascade the survey invite email, including via:

• The Hospital at Home Programme Board.
• The existing H@H adult network, managed by HIS.
• The existing H@H Paediatric and Neonatal network, managed by HIS.
• The Scottish Government’s General Practice Policy Division.
• The Scottish Government’s Primary Care Out-of-Hours Policy Unit.
• The Scottish Government’s Allied Health Professionals Policy Unit.
• Existing H@H Heart Failure pathway network.
• Existing H@H Respiratory pathway network.
• Existing H@H OPAT pathway network.
• Stakeholder colleagues in the Scottish Ambulance Service.
• Stakeholder colleagues from the Royal Colleague of Emergency Medicine.

Recipients of the invite email were asked to take part in the survey, and to also share it within their respective areas of work. The first invite emails were sent on 24th November 2025, and a reminder was sent through most avenues during December 2025. This method aimed to reach a broad cross section of healthcare professionals across a wide range of geographies. Due to the sampling strategy, the total number of individuals who received the workforce survey is unknown, and therefore a response rate cannot be calculated.

Workforce survey response and profile

The following sections provide information on the workforce survey respondent cohort, including the volume of responses and contextual information on respondents’ role, work area (e.g. General Practice, Emergency Medicine, etc.), Health Board, and H@H pathway involvement. The purpose of providing this information is to support interpretation of the findings and any sub-cohort divergences, not to provide a reflection of the distribution or nature of the H@H service workforce across Scotland. The survey sample should not be taken to be representative of the wider health and care workforce delivering H@H in Scotland.

A total of 236 respondents completed the workforce survey during the period from November 24^th 2025 to January 19^th 2026. As set out above, respondents were split into two cohorts: those delivering H@H care, and those referring patients to H@H. The responses received from each cohort were:

• 100 respondents delivering H@H services.
• 136 respondents referring patients into H@H services.

Across the workforce survey as a whole, responses were received from 13 territorial Health Boards (representing all Boards in Scotland except NHS Orkney), plus a small number (n=7) from respondents not based in a territorial Health Board^[24]. However, there were no responses from deliverers in NHS Borders, and no responses from referrers in NHS Western Isles.

Workforce survey delivery staff profile

Table A1 shows the profile of deliverers by role. The most commonly selected roles were nurse (23%), followed by Consultant (19%) and Advanced Nurse Practitioner (16%).

Respondents who delivered H@H (‘deliverers’) described their area of work as either a ‘Hospital at Home Team’ (65%) or a ‘Specialist team which includes delivering Hospital at Home care’ (33%). Respondents in a ‘Hospital at Home Team’ are likely to deliver only H@H services, whereas those in a specialist team are likely to spend a proportion of their time delivering H@H and a proportion elsewhere.

Delivery respondents were asked which Hospital at Home pathways they were involved in delivering, with the option to select all that applied from the existing pathways: Acute/Older adults, Respiratory, Heart Failure, OPAT, Paediatrics, Neonatal. While just over half (53%) of deliverers said they delivered a single pathway, 47% reported that they delivered two or more pathways. Those in H@H teams (48%) were more likely than those in specialist teams (24%) to report involvement in the delivery of multiple pathways^[26]. It should be noted that the presentation of these sample profile findings are not intended to offer an assessment of the national landscape of service delivery, but rather to provide context in relation to the survey respondent sample.

Table A2 shows that three in four deliverers said they were involved in delivering the Acute/Older adults’ pathway (either solely or in combination with other pathways). Neonatal and Paediatric pathways were the least common.

Note: Totals exceed 100% because respondents could be delivering more than one pathway.

It should be noted that the sample was a self-selecting sample and should not be considered as representative or a reflection of service provision across Scotland. Some notable variations in responses by pathway involvement and geographical area emerged, including:

• None of the ten deliverers in Greater Glasgow and Clyde said they delivered the Acute/Older adults or the Respiratory pathway.
• The Paediatric pathway was only selected by respondents in NHS Lanarkshire and NHS Tayside.
• The Neonatal pathway was only selected by respondents in NHS Greater Glasgow & Clyde and NHS Tayside.

• Those in a H@H team were more likely than those in a Specialist team which delivers Hospital at Home care to be involved in the delivery of:

  • the Acute/Older adults pathway (88% compared to 52%).
  • the Respiratory pathway (60% compared to 24%).

Deliverers were asked, in relation to the Hospital at Home pathway they were involved in delivering in their respective Health Board areas, how they would describe the current status of the service. As nearly half of respondents said they were involved in delivering more than one pathway, in many cases it was not clear which pathway their responses related to. Overall, however:

• Nearly half (48%) described their pathway as well established.
• 29% said it was reasonably well established.
• 14% said it had recently been introduced and was receiving referrals.
• 9% said that their pathway was under development but not yet launched or receiving referrals.

Those involved in delivering adult pathways were more likely to consider their pathway(s) as well established, whereas those involved in delivering child pathways were more likely to consider their pathway(s) as less established or under development.

Table A3 shows the profile of the referrer respondents by role.

Workforce survey referral staff profile

As set out in the Evaluation scope and definitions section at the beginning of the report, an admission to H@H can be as an alternative to a hospital admission or following a hospital stay as part of an early supported discharge. For context, GP referrers would be more likely to refer patients to a H@H service on an admission avoidance basis (although not exclusively). Conversely, Emergency Department/hospital-based staff (including Consultants, who made up 13% of the sample), would be more likely to refer into H@H as part of an early supported discharge.

Workforce survey data analysis

Workforce survey data were exported to SPSS for analysis. Data were cleaned, which included adding variable labels (question text) and transforming response data from text to numerically coded values with labels. The ‘I don’t know’ responses were coded as ‘missing’ which excludes them from analysis. Open-ended survey responses were coded by theme, with the creation of multiple new variables to categorise text according to emerging themes. New variables were created which aggregated responses (e.g. ‘agree’ and ‘strongly agree’). Frequencies (counts and percentages) were run for all closed question responses.

Crosstabulations with chi square tests to test for significant differences were run for:

• All variables in the survey of deliverers and the survey of referrers (disaggregate and aggregate) by role - these did not show any significant differences.
• Level of promotion of the service (strong promotion; some/limited/very little or none) with level of agreement (strongly agree or agree; other response) with statements on staffing, resourcing and operation (survey of deliverers).
• All variables in the survey of referrers (disaggregate and aggregate) by area of work (General Practice; Scottish Ambulance Service; hospital inpatient emergency medicine; community/district nursing; primary care out-of-hours; ED/MIU based emergency medicine; Flow Navigation Centre).
• Perceived ease of referral (mostly or almost always straightforward; other response) with other variables on views of H@H (aggregate and disaggregate).

Where crosstabulations were conducted, only differences between groups which are statistically significant (p<0.05) are reported.

Health and Social Care workforce focus groups and interviews

Background and methodology

Qualitative research with relevant NHS workforce included online focus group sessions and individual/paired interviews.

Separate topic guides were developed for sessions with referral and delivery staff, which were also used for corresponding interviews. Materials were developed collaboratively by the research team and Scottish Government, and were designed largely to fit around the overall aims for the research and the H@H logic model. The majority of questions were process focused, although some did allow for early wider reflections on the impact and future of H@H for both patients and staff. Information sheets and privacy notices were also shared for comment and review, and copies of these were made available to all those invited to take part.

All of those who had indicated as part of their workforce survey response that they would be willing to be contacted for follow-up were invited to attend one of a series of scheduled focus group sessions, or to opt in to a one-to-one interview (if the pre-arranged sessions were not suitable). As the number of individuals who initially responded to this invitation was lower than anticipated, several booster activities were also undertaken. This included invitations to take part being cascaded via various relevant professional networks. A separate online opt in proforma was developed for this purpose.

Focus group and interview response and profile

A total of six online focus group sessions were run - three with delivery staff (attended by 13 participants in total) and three with referral staff (attended by six participants in total). Each session was facilitated by two researchers and questions were shown to participants on screen during sessions to aid discussions. All sessions lasted for one hour.

A total of eight interviews were also carried out (capturing views of nine staff). This included seven delivery staff and two referral staff. These were carried out either by telephone or online and lasted an average of just over 30 minutes. The 28 delivery and referral staff who took part in the qualitative engagement represented a range of different professional roles/functions. Participants included Consultants, Advanced Nurse Practitioners, GPs/GP Partners, Managers, Allied Health Professionals, Resident Doctors and those who described their role as ‘Other’. There was also representation from 10 Health Boards, but none from NHS Borders, NHS Shetland, NHS Orkney or NHS Western Isles.

Focus group and interview data analysis

All focus group and interview sessions were digitally voice recorded with permission from participants and were transcribed verbatim to allow for iterative thematic analysis, to expose underlying themes that emerged naturally from the data.

All transcripts were read and initial categories within the data were identified (i.e. clusters of data were grouped and given descriptive labels illustrating key concepts). Coding of clustered data was carried out wherein categories were given analytical rather than descriptive labels and were further broken down where possible, to highlight nuances in the feedback given. The researchers undertook constant comparative analysis whereby, as more data were analysed, new categories were isolated and existing categories were broadened or made more sophisticated to accommodate new data samples (e.g. sub-categories). Any negative or outlier cases were identified and the analysis continued until saturation had been reached (i.e. all new possible extractions from the data were exhausted and no new themes were identified). Anonymised verbatim quotes were selected for key themes for inclusion in the report to highlight key emergent themes.