Development of the support needs after ICU (SNAC) questionnaire

Abstract Aims To develop a questionnaire to identify Intensive Care survivor needs at key transitions during the recovery process, and assess its validity and reliability in a group of ICU survivors. Methods Development of the Support Needs After ICU (SNAC) questionnaire was based on a systematic scoping review, and analysis of patient interviews (n = 22). Face and content validity were assessed by service users (n = 12) and an expert panel of healthcare professionals (n = 6). A pilot survey among 200 ICU survivors assessed recruitment at one of five different stages after ICU discharge [(1) in hospital, (2) < 6 weeks, (3) 7 weeks to 6 months, (4) 7 to 12 months, or (5) 12 to 24 months post‐hospital discharge]; to assess reliability of the SNAC questionnaire; and to conduct exploratory data analysis. Reliability was determined using Cronbach's alpha for internal consistency; intraclass correlation coefficients for test–retest reliability. We explored correlations with sociodemographic variables using Pearson's correlation coefficient; differences between questionnaire scores and patient demographics using one‐way ANOVA. Results The SNAC questionnaire consisted of 32 items that assessed five categories of support needs (informational, emotional, instrumental [e.g. practical physical help, provision of equipment or training], appraisal [e.g. clinician feedback on recovery] and spiritual needs). ICU survivors were recruited from Northern Ireland, England and Scotland. From a total of 375 questionnaires distributed, 202 (54%) were returned. The questionnaire had high internal consistency (0.97) and high test–retest reliability (r = 0.8) with subcategories ranging from 0.3 to 0.9. Conclusions The SNAC questionnaire appears to be a comprehensive, valid, and reliable questionnaire. Further research will enable more robust examination of its properties e.g. factor analysis, and establish its utility in identifying whether patients' support needs evolve over time. Relevance to clinical practice The SNAC questionnaire has the potential to be used to identify ICU survivors' needs and inform post‐hospital support services.


| BACKGROUND
The need for support throughout the post-ICU recovery process has long been identified. [1][2][3][4] A key priority for the intensive care community is determining how and when to support ICU survivors and their families/carers. 5,6 Guidelines recommend that individualized assessment, rehabilitation, and multidisciplinary follow-up services should be provided, 7,8 but there is little consensus on the components, individualization, timing, mode of delivery (e.g. face to face, telephone or online), and duration of such services. [9][10][11][12][13][14] Furthermore, there are no standardized, validated tools to evaluate the extent to which these services, where they exist, meet patients' needs. We aimed to address this gap, by developing a support needs questionnaire for intensive care survivors which could be used to assess patient identified needs at key transitions during the recovery process.

| Support needs
Support needs have been defined as "the additional help some adults need in order that they can live in the best way they can, despite any illness or disability they might have." 15 Support needs questionnaires have been developed for other populations and these can help to shape service provision and co-ordination, and enhance a patient centred approach to care. [16][17][18][19][20][21] The Social Support Needs framework developed by House categorized needs as informational, emotional, instrumental (e.g. provision of practical physical help, training or equipment), and appraisal (e.g. reassurance, clinician feedback on recovery), which may be short-or long-term and/or evolve across the recovery trajectory. 2,15,22 We previously conducted a scoping review of the qualitative literature on ICU survivors' support needs 15 and identified the Timing it Right framework (TIR) 23,24 as a useful and relevant means of capturing ICU survivors' support needs during the different stages of recovery.
The TIR framework was initially developed for use among family members of stroke survivors, 23,24 and it provides a practical, timebased framework in which survivors' needs during recovery can be explored. The TIR framework has also been used among ICU survivors 4 to describe key transition phases across the recovery continuum: from the event in ICU, to stabilization on the hospital ward, preparation for hospital discharge, the early phase at home, and longer-term recovery (adaptation phase). 4 This study is framed around the recovery stages that align to the TIR framework (excluding the event in ICU). 4

| METHODS
The study was conducted in two stages: 1. questionnaire development and assessment of validity; and 2. a pilot survey to assess recruitment of ICU survivors, reliability of SNAC, and exploratory analysis of data ( Figure 1). Ethical approval was granted from the Office for Research Ethics Northern Ireland 17/NI/0236 and research governance for individual collaborating sites was obtained. Completed and returned questionnaires were accepted as informed consent.
Questionnaires were coded to ensure confidentiality and all data were stored on password protected computers in accordance with the Ulster University research procedures.

| Stage 1. Questionnaire development and assessment of validity
The approach to developing and establishing face and content validity of the SNAC questionnaire followed key recommendations for developing questionnaires. [25][26][27][28][29] What is known about the subject • A key priority for the intensive care community is to identify how and when to support ICU survivors at key transitions in care.

What this paper contributes
• This new Support Needs After ICU (SNAC) questionnaire has the potential to inform the development of support services for ICU survivors.
• Future research using the SNAC questionnaire following the same group of patients across their recovery continuum would identify whether patients' support needs evolve over time.

| Item generation
Items were generated from three sources: detailed findings from a longitudinal qualitative study of ICU survivors' support needs during the year post-ICU discharge 30 ; a published scoping review of 32 qualitative studies 15 ; and iterative consensus meetings among the multidisciplinary research team.
The longitudinal qualitative study was undertaken previously by a co-author (PR) among adult ICU survivors who were mechanically ventilated for ≥2 days and discharged from one of two Scottish ICUs. 30 Interviews were undertaken at four time points: prior to hospital discharge, and at 4 to 6 weeks, 6, and 12 months following ICU discharge. These time points align to the TIR recovery phases including stabilisation and preparation for hospital discharge, and the early and longer-term home recovery. Using thematic analysis, data were coded with reference to patient support needs identified in the TIR framework. Findings showed that support needs commonly included making sense of the critical illness experience (including illness severity, amnesia, dreams, hallucinations, flashbacks); making sense of physical, psychological, and emotional frailty; a perceived lack of understanding among clinicians; social isolation following hospital discharge; vulnerability; concern for family members; and the socioeconomic burden of critical illness. 30 The scoping review included published qualitative research from 2000 to 2017 reporting on adult ICU survivors' support needs. 15 Two reviewers (JK, PR) independently screened, extracted and thematically analysed the research themes with reference to the support needs identified in the TIR framework. Informational, emotional, instrumental, appraisal, and also spiritual support needs were identified, and the nature and intensity of need differed when mapped against the timepoints of the TIR framework. 15 The initial questionnaire (version 1.0, n = 31 items) was devel-

| Data analysis
Feedback from the ICU service users and health professionals was collated and reviewed by the research team. The team discussed items that needed changed, added, reduced or wording altered and consensus was reached regarding any amendments. This process was repeated following further review by the additional ICU service users.
For congruency, the average percentage agreement between service users for each question in the SNAC questionnaire was calculated.
F I G U R E 1 Summary of the SNAC Questionnaire development process The Drivel Defence Index 33 provides an automated calculation of the number of words in a statement or sentence; any statement with more than 20 words was reviewed and amended.
2.2 | Stage 2. Pilot survey to assess recruitment of ICU survivors at key stages of recovery, reliability of the SNAC questionnaire, and conduct exploratory analysis of data

| Participants and setting
We recruited general ICU survivors from four UK hospitals and from a UK patient support group, ICU Steps (www.ICUsteps.co.uk). We included participants at five time points aligned to the TIR framework: (1) on the hospital ward; (2) discharged from hospital within 4 to 6 weeks; (3) at home between 7 weeks to 6 months; (4) between 7 to 12 months; (5) or 12 to 24 months.
Questionnaires were administered by post, by dedicated research staff in each setting. Inclusion criteria comprised adults over 18 years old with an ICU admission in the previous 12 months requiring mechanical ventilation for ≥48 hours. We excluded survivors requiring palliative care, planned specialist support pathway (e.g. liver transplant), with a neurological event (e.g. head injury or neurodegenerative condition), or if they declined or were unable to give consent.
The return of the completed questionnaire was accepted as consent.

| Sample size
A formal sample size calculation is not required during pilot testing. 34 However, we aimed to recruit 200 participants from the four sites to obtain preliminary data for the iterative refinement of the questionnaire, to ensure sufficient data to assess adequacy of our analysis procedures, and to inform ability to recruit ICU survivors. The sample included representation from across the continuum of recovery after ICU i.e. ward, 4 to 6 weeks post discharge, 6 months, and 12 months.
We estimated 50 participants of the total sample would be acceptable to complete the questionnaire on a second occasion for the test retest reliability. This was based on recommendations that a minimum of 22 participants would be required to detect a medium sized intraclass correlation coefficient (ICC) of 0.5. 31 We over-inflated the number to ensure that any smaller ICCs were detectable.

| Recruitment procedure
Patients were identified from ICU registers/databases by dedicated staff at each of the participating sites, or via email from ICU Steps support group (for patients 6 or 12 months after an ICU stay). Those meeting inclusion criteria were provided with an envelope (if in hospital) or posted (if discharged) containing the study information and the SNAC questionnaire.
We conducted test/retest reliability by including a postcard asking participants to consider completing the questionnaire on a second occasion. On return of the postcard indicating consent, participants were posted a second copy of the questionnaire 2 weeks later. 35   Respondents were either still in hospital (n = 19, 9.5%); or discharged from hospital less than 6 weeks ago (n = 33, 16.5%); discharged from ICU between 7-weeks and 6-months (n = 71, 36%); 7 to 12 months (n = 51, 25.5%); or 1 to 2 years (n = 26, 13%). Respondent characteristics are summarized in Table 1. There were less than 5% missing responses from the potential total responses for the SNAC questionnaire. Given this low rate, the chances that these data had a major influence on our findings is minimal, 37 indicating relevance and perceived importance of the items included. One statement pertaining to need for "information about returning to work" was missed by a  Table 1). Table 2 shows the mean (SD) overall scores for the total SNAC (range 32-160) and each category; information (range 12-60); emotional (range 9-45); instrumental (range 6-30); appraisal (range 4-20); and spiritual (range 1-5) within each of the five timepoints. Figure 2 shows the proportion of patients with support needs (i.e. those who responded that they were undecided, or agreed, or strongly agreed that they had specific support needs 25

| DISCUSSION
A key priority for the intensive care community is to identify how and when to support ICU survivors at key transitions in care, and throughout the recovery process. We used an iterative and robust process to develop a patient-centred questionnaire to identify patient needs T A B L E 2 Mean score for the SNAC questionnaire and for each needs category at each recovery timepoint after ICU discharge

| Strengths and limitations
This study has a number of strengths. First, the use of an existing theoretical framework to identify support needs and key recovery stages after ICU. 4,15,[22][23][24]30 Second, a robust and iterative process that included views from patients with experience of a stay in ICU throughout the development process. [25][26][27][28][29] Third, the questionnaire facilitated patients to identify their own support needs 48 The study also had a number of limitations. First, the response rate was 54%. It is, therefore, not possible to ascertain if nonresponders had higher, or no, support needs; it was too burdensome to complete; or they did not want to participate in research. Unlike general population surveys where response rates less than 70% are considered at risk of bias, 49  This framework helps researchers consider which ethnic groups should be included to ensure results are widely applicable, and the challenges in making this possible. 52 Fourth, the timeframes may be considered wide, e.g. the stage between 7 weeks and 6 months, as patients' needs may change during this time. However, the timelines for assessment of the 5 stages of recovery were broadly aligned to the TIR framework which we had used for the study from the outset.

| CONCLUSION
The SNAC questionnaire appears to be a comprehensive, valid, and reliable questionnaire that could help to identify ICU survivors' support needs. Although there are some limitations of the pilot survey, the SNAC questionnaire has the potential to discover which support services need to be available for ICU survivors. Subsequent research will enable a more robust examination of its psychometric properties e.g. factor analysis, and establish its utility in identifying whether patients' support needs evolve over time. Future work may also need to explore the extent to which patients feel able to engage with support services even when these are available.

ETHICS STATEMENT
Ethical approval was granted from the Office for Research Ethics Northern Ireland (Reference 17/NI/0236) and research governance for individual collaborating sites was obtained.

PATIENT CONSENT STATEMENT
The return of the completed questionnaire will be accepted as consent. Additionally, we conducted test/retest reliability by including a postcard asking participants to consider completing the questionnaire on a second occasion. On return of the postcard indicating consent, participants were posted a second copy of the questionnaire 2 weeks later.

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.