Measurement properties of scales assessing new graduate nurses' clinical competence: A systematic review of psychometric properties

Background: New graduate nurses' competence is a concern for all healthcare organizations. Previous reports show heterogeneous levels of competency amongst them. As a positive association between competency and quality of care in clinical settings has been suggested, it is essential for researchers and clinicians to select valid, reliable, and responsive scales to assess new nurses' competence. However, a systematic evaluation of the measurement properties of scales measuring new nurses' competence had yet to be published.

Objective: To analyse, evaluate and synthesize the measurement properties of scales used to assess new nurses' clinical competence.

Design: A systematic psychometric review based on the COnsensus-based Standards for the selection of health Measurement INstruments (COSMIN) methods.

Data sources: The search strategy included a combination of keywords and thesaurus terms related to new graduate nurses, clinical competence, and competence assessment. Five databases were searched: Embase, CINAHL, MEDLINE, PsycINFO and Web of Science. The search was limited to full-text papers published in English or French, from 2010 to 2019.

Review methods: Two independent reviewers screened eligible papers, extracted data related to validity, reliability, and responsiveness of each scale, and evaluated the quality of their measurement properties as well as risk of bias in their psychometric evaluation. Divergences were solved through discussion.

Results: Ten scales were included: eight original scales, one culturally adapted and one modified. Of these scales, eight were developed or adapted in the 2010s decade and the other two scales were developed earlier. Most scales are divided into 6 to 8 subscales and use an adjectival scale with either 4, 5 or 7 points. The content validity study of all scales in this review was deemed to be doubtful or inadequate quality. Reliability was almost exclusively assessed by calculating the internal consistency with Cronbach's alpha coefficient which gives no information on equivalence or stability of the measure. Responsiveness was never properly assessed in the reviewed studies.

Conclusions: There is little evidence on the measurement properties for each scale regarding their validity and reliability; responsiveness was not assessed for any scale. Every scale evaluated in this review had different characteristics (length, subscales, response options). Therefore, selection of the most appropriate scale depends on the context and purpose of the assessment. Prospero registration number: CRD42018109711 Tweetable Abstract: Systematic review of scales measuring new nurses' competence: we must do better and conduct more validity/reliability testing of existing scales.