SS-INQ Information Needs Questionnaire in Sjögren's Syndrome

This is a PhD student project that will be undertaken in 2 distinct phases.

Phase 1:

Determination of Informational Needs (Months 0- 9)

• Qualitative interviews with patients with SS
• Interview transcriptions and thematic analysis

Subjects The perspective of the patient is critical in the determination of the informational needs of patients with SS. Qualitative interviews will be conducted patients with SS. Due to the depth of rich data that can be derived from focus groups we plan on using this method of qualitative interview.

Streiner and Norman proposed that each focus groups would contain 6 to 12 patients. Approximately 40-50 patients with SS will be recruited for the qualitative interviews. These patients will be identified from the Oral Medicine clinics of the UCLH Eastman Dental Hospital. Inclusion criteria will include:

• Adult patients with a diagnosis of primary SS as per the 2002 American-European Consensus Group (AECG) classification criteria, or
• Adult patients with a diagnosis of primary SS as per the 2016 American College of Rheumatology/European League Against Rheumatism classification criteria
• Adult patients who are able to consent to being involved in the study
• Willingness to participate in focus groups
• Fluency of the English language to allow participation in focus groups

Research methods:

These focus groups will take the form of semi-structured interviews, using a topic guide as a general framework during the interview. The topic guide will consist of a list of topics of interest, developed by -

• informal interviews / discussions with patients and clinicians
• clinical observation The order of the items on the list will not be fixed and, therefore, will allow for flow of the conversation.

With the qualitative interviews data collection is carried out and analysed in stages. This process continues until data saturation has been achieved. Data saturation is deemed to have been achieved when no new information or themes emerge. Once data saturation has been achieved the digital recordings of the focus groups will then be transcribed to allow analysis of the data.

Data analysis:

Although multiple approaches to qualitative data analysis exist, three are more commonly used, namely thematic analysis, grounded theory and framework analysis. Thematic analysis will be used and the emergent themes will allow for the determination of the informational needs of patients with SS.

Phase 2 Development of the informational need instrument for SS [SS-INQ] (Months 9-32)

• Adaptation of TINQ-BC for use in patients with SS - generation of relevant questions using themes from Phase 1 qualitative study (removal of those solely related to breast cancer from TINQ-BC) by the expert group
• Pilot testing of SS-INQ for content and readability will be done via focus groups whilst structural validity will be explored using factor analysis
• Testing of SS-INQ for reliability - internal consistency reliability and test-retest reliability Subjects (including justification for sample size) The number of experts involved in item generation of a new instrument varies widely in the literature, from as few a 3 to as many as 24. With no established consensus on the critical number of persons involved it is our intention to establish a 5 to 7 person expert group, consisting of Rheumatologists, Oral Medicine Clinicians and Ophthalmologists, to be involved in the generation of questionnaire items.

A previously mentioned we propose to hold focus groups with 6 to 12 patients per group. Approximately 12 to 24 patients with SS will be recruited for content validity and readability.

Tinsley and Tinsley (48) suggest a ratio of about 5 - 10 subjects per question in the instrument up to 300 subjects when carrying out factor analysis. TINQ-BC consists of 52 items, therefore assuming that the SS-INQ will consist of a similar number of items a minimum of 260 patients will be required to determine the structural validity of the newly developed instrument using factor analysis.

The COnsensus-based Standards for the selection of health Measurement Instruments (COSMIN) checklist can be used to determine the methodological quality of measurement tools. According to Mokkink et al 50-100 patients to achieve a good to excellent quality assessment rating when determining internal consistency reliability and test-retest reliability.

Research methods and data analysis:

Due to the importance of an underlying structural framework in the development of a questionnaire we plan to adapt an existing informational needs questionnaire, which was developed used Lazarus and Folkman's Theory of Stress and Coping conceptual framework. TINQ-BC was developed in the late 1990s due to a lack of knowledge about the amount and type of information patients with a recent diagnosis of breast cancer may require. In a recent review TINQ- BC was deemed to be a tool that should be used as a model in the development of future informational needs questionnaires. The themes derived from the qualitative interviews will be used to populate the instrument ensuring it is specific to SS and reflective of the patients' informational needs.

The SS-INQ instrument will then be tested for content and structural validity, readability and reliability as follows:

• Content validity and readability: content validity refers to the extent to which the items in the questionnaire reflect the facets of the domain or construct they are proposed to represent. Each facet of the domain should therefore be represented by at least one item in the questionnaire.
• Structural validity: factor analysis will be conducted to determine the dimensionality of the instrument. The goal of factor analysis, according to Kin and Mueller, is 'to represent a set of variables in terms of a smaller number of hypothetical variables'.
• Internal consistency reliability: defined as 'the precision of a scale, based on the homogeneity of the scale's items at one point in time', will be determined by calculating Cronbach's Alpha coefficient. Alpha values can range from 0.0 to 1.0. It refers to the degree to which statements in SSINQ correlate with each other. According to De Vellis and Streiner and Norman levels between 0.7 and 0.9 are desirable.
• Test-retest Reliability: Test-retest reliability measures reliability of the instrument selected over time. In general instruments are administered at two points in time. A period of 12 to 14 days is generally considered an acceptable retest interval. Anastasi and Urbina suggested that the value of the intraclass correlation coefficient (ICC) should generally be > 0.8 . We will administer the chosen instrument at two time points 14 days apart. The ICC will then be calculated.

Final analysis 32-36 months.