The use of coercion in health care is ethically problematic and challenge the fundamental health care principle of respect for patient autonomy (1, 2). All over the world there are initiatives to minimize its use (3–6). Health care professionals need to critically reflect upon and morally justify each use of coercive interventions (7, 8). Several studies have shown considerable variation in use of coercive measures both, in one country and between different countries (9–11). These differences are not yet fully explained (12–14). Based on differences in the use of coercive practices among different countries, regions, and hospitals, some of the variation can be attributed to differences in staff attitudes to the use of coercion (15). Attitudes can be defined as a psychological tendency that is expressed through evaluating an entity with a normative degree of either positivity or negativity, based on experience (16). Attitudes do influence behavior, but the connection between attitudes and behavior is complex, and may depend on situational factors. The connection has not been fully mapped yet, and the relationship may also depend on the subject of the attitude (17).

In recent years, attitudes to the use of coercive interventions in mental health care have evolved, with increased focus on user participation, respect for autonomy, and human rights (6). Differences in staff attitudes to the use of coercion may explain why some wards and hospitals have attempted to reduce the use of coercion, while others have not made the same effort (4, 18). Staff attitudes to coercion may also influence the amount of coercion used and reveal the reasons for using coercion in treatment and the dynamics involved.

It is therefore important to have a validated questionnaire for assessing staff attitudes to coercion in mental health care. In 2008, Husum et al. developed and published the Staff Attitude toward Coercion Scale (SACS) for this purpose (19). SACS measures staff attitudes toward use of coercive practices in mental health care. SACS was developed as a short 15-item questionnaire with normative attitudes toward use of coercion. It consists of statements about the use of coercion, about how the participant thinks about it, and whether the participant considers coercive interventions necessary. Using factor analysis, the questionnaire divided staff attitudes to coercion into three groups: the view that use of coercion may offend (critical attitude); the view that coercion is necessary for care and security reasons (pragmatic attitude); and the view of coercion as a valid form of treatment (positive attitude). These items are scored on a five-point Likert scale, from 1 = strongly disagree to 5 = strongly agree.

The SACS is to our knowledge the only instrument measuring staff attitudes to coercion, and the instrument has now been used worldwide and translated into several languages, including German, Polish, Chinese, Japanese, Italian, Turkish and Arabic, indicating a potential for cross-cultural applicability. The questionnaire has also been used in some populations with other participants, like patients and caregivers. However, to the best of our knowledge, no systematic review has been performed to assess the measurement properties of the scale. To date, there has been no attempt to examine the results of its use in practice and research, and no meta-analysis has been done.

This review aims to gather results relevant for measurement evaluation of the SACS questionnaire. Assessing measurement characteristics is essential for comparing results from different countries and populations. In particular, the aim of this study was to conduct a systematic review of the measurement properties of SACS using the COSMIN Risk of Bias checklist (20–23). The following review questions were considered.

This systematic review was carried out following the “COnsensus-based Standards for the selection of health Measurement Instruments” (COSMIN) (20, 22, 23) and the “Preferred Reporting Items for Systematic reviews and Meta-Analyses” (PRISMA) guidelines (24).

The review protocol was registered with PROSPERO (Record ID = CRD42021239284).

The SACS has been used in many studies and countries since its development in 2008. This indicates concern in many cultural settings about use of coercion in mental health care. However, to the best of our knowledge, this is the first systematic review of the measurement properties of the SACS.

Structural validity and internal consistency were the most frequently reported measurement characteristics in the identified studies. All studies reported adequate internal consistency. Most studies replicated the original three-factor structure and the correlation between the factors were medium to low as expected. Merging the model into one factor has been considered, but the measurement findings in this review suggest that the three-factor is better fit to the available data. Keeping the three-factor model is supported by another study who found that professionals could be divided in three groups concerning their thoughts about use of seclusion in mental health care. The authors of this study identified three types of professionals: Transformers, Doubters, and Maintainers (36).

While the available data support the SACS as being psychometrically sound, some important data are missing. For example, there are no measurements across more extended periods of time (37), which could be used to assess the sensitivity to change and stability of the SACS. Two test-retests within a 3-week interval indicate that SACS score may vary over time in the same individuals indicating that the SACS may assess dynamic attitudes likely to fluctuate even within relative short time periods.

Testing of criterion validity by comparing SACS to a gold standard is lacking because there are no other instruments on staff attitude to coercion to compare with. There is also very limited research on the relationship between staff attitudes and the use of coercion. Several studies have suggested that there is a relationship between staff attitudes toward use of coercion and actual use of coercive interventions (8, 26, 35, 38–43).

Other studies have found important differences between explicit and implicit attitudes (44). Explicit attitudes are attitudes that the individual himself is aware of, while implicit attitudes are those attitudes that one is not aware of, but can be measured indirectly through, for example, autonomous reactions. The most common approach to measuring explicit attitudes is self-reporting as these are attitudes that the subject is self-aware of. The SACS questionnaire measures only explicit attitudes. One of the studies included in this review had however studied both explicit and implicit attitudes (45). This is also a topic that should be investigated more thoroughly in future studies.

As shown in Table 3, the reporting of measurement properties except structural validity and internal consistency are so far almost non-existent in SACS studies. A major reason for this may be those analyses of factor structure and the internal consistency of scales and subscales are well established and fairly easy to do, while several of the other properties have been more defined as standards more recently and are less established. Some of these also require more demanding design, data collections and data analyses. The detailed and complicated criteria in the COSMIN rating instructions also makes it difficult to have simple descriptions in a table explaining what the ratings mean. Some criteria in the COSMIN Risk of Bias checklist have also been debated, like rating properties of an instrument based on whether the instrument confirms a hypothesis in a study, as it is impossible to assess in one study both the treatment effect and the responsiveness of an instrument (46).

A challenge in using the COSMIN method for this systematic review is also that the COSMIN taxonomy is originally developed for the assessment of patient-outcome measures. The SACS measures staff attitudes, which is another kind of phenomenon. In general, the scientific study of staff attitudes in mental health care seems to be scarce, and this field needs to be methodologically developed.

The SACS is, to our best of knowledge, the only questionnaire measuring staff's attitudes to the use of coercive interventions in mental health services. It is used widely, which demonstrates the need for such a tool. The widespread use also indicates that the tool is perceived as feasible and useful.

The assessment found evidence for adequate validity and internal consistency of the SACS. However, there were very limited support for other important measurement qualities such as reliability, criterion validity and measurement error.

Future research should focus on the stability of these attitudes, whether they are amendable by interventions and the relationship between staff attitudes to coercive interventions and the actual use of coercion. Another related topic is to more thoroughly investigate the relationship between staff's explicit and implicit attitudes toward use of coercion. Further future research could investigate formation of staff attitudes. Staff attitudes to the use of coercive practices may also influence the staff's willingness to engage in projects to reduce the use of coercive interventions. Another possible topic for research is to investigate barriers to engage in projects aimed at reducing use of coercive practices in mental health care.

Until future studies have evaluated more measurement properties of SACS, users of SACS must interpret the results based on the current knowledge of its properties.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

All authors have been involved in the process of assessing the studies for inclusion, analyzing the results, writing article, contributed to the article, and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.