Translation and cross-cultural adaptation to Brazilian Portuguese of two brief screening tools for youth at risk of psychosis: the Prodromal Questionnaire (PQ-16) and the PRIME-Screen

Abstract Introduction Prodromal characteristics of psychosis have been described for more than a century. Over the last three decades, a variety of studies have proposed methods to prospectively identify individuals (and youth in particular) who are at high risk of developing a psychotic disorder. These studies have validated various screening instruments and made them available in several languages. Here, we describe the translation into Brazilian Portuguese and cross-cultural adaptation of two such screening tools – the Prodromal Questionnaire-16 (PQ-16) and the Prevention through Risk Identification, Management, and Education (PRIME)-Screen. Method Two bilingual native speakers of Brazilian Portuguese translated the questionnaires from English. A native English speaker then performed back-translations into English. These back-translated versions were submitted to the original authors. They provided feedback and later approved the final versions. Results After translation and cross-cultural adaptation, no items needed to be changed in the adapted PQ-16 and four items were revised in the PRIME-Screen. After the peer-review process, we included two suggestions in the PQ-16 to facilitate use of the tool in our cultural and social contexts. The PRIME-Screen did not need further changes. Conclusion These new instruments can help screen Brazilian Portuguese-speaking patients who are at risk of psychosis in primary care.


Introduction
Psychotic disorders are the ninth-leading cause of global health problems. [1][2][3] People who suffer from these disorders often face delayed identification and treatment of their condition. 4 In addition, patients often report subthreshold symptoms that are associated with mild to moderate functional impairment before their first episode of psychosis (FEP), which typically occur from weeks to months before the full-blown psychosis.
Prospective assessment of subsyndromal symptoms enabled researchers to identify patients who were at higher risk of developing psychosis, which led to definition of the at-risk mental states (ARMS). Previous studies have developed brief questionnaires to screen for ARMS, but to date few have been available for lowand middle-income countries.
Some commonly reported subthreshold psychotic symptoms that characterize the ARMS syndrome include thought disorders, altered beliefs, and changes in perception and speech. In the early 1990s, an Australian group led by McGorry and Yung developed a semistructured clinical interview to prospectively assess young people who presented such symptoms. They called their interview the Comprehensive Assessment of At-Risk Mental States (CAARMS). 5,6 A few years later, an American group called Prevention through Risk Identification, Management, and Education (PRIME) developed their Structured Interview for Psychosis-Risk Syndrome (SIPS). This tool aimed to evaluate the same phenotype. 7 These tools have been proven to reliably identify people who are in ARMS and have opened the door for a new field of study called early intervention in psychiatry. 8 Comparative studies have found that both instruments perform similarly and both were able to identify people who are likely to transition from ARMS to full psychosis after two years. 9 Early intervention has already become government policy in several developed countries (e.g., Australia and United Kingdom). However, implementing early intervention in clinical practice has proven to be a challenge. For instance, it is difficult to identify ARMS subjects correctly. To conduct CAARMS and SIPS interviews, clinicians require extensive training, specialized staff, and considerable time to administer the interviews. Consequently, using these interviews to identify people who are at risk of developing psychosis in daily clinical practice is unfeasible, especially in low-resource settings. Over time, these challenges led to development of self-reported screening tools such as the Prodromal Questionnaire (PQ) and the PRIME-Screen. 10 The long version of the PQ has been proven to have adequate psychometric properties and a fair degree of reliability. 11,12 It has also been translated into Portuguese. 13 Unfortunately, this 92-item questionnaire is very long and therefore not appropriate as a screening tool for use in non-specialized settings. A Dutch group recently tested a condensed version of the tool -the PQ-16 -and demonstrated that it is a reliable and much more time-effective tool. 14 Similarly, the North-American group developed the PRIME-Screen, a brief questionnaire based on the SIPS. 15 The PRIME-Screen is more specific and sensitive than most other screening tools. 16 Consequently, the PRIME-Screen has been widely adopted in studies of early detection of psychosis that use a two-stage screening process, comprising an initial brief screening procedure and a subsequent clinical interview. 17 To date, there are no brief screening instruments for ARMS available in Brazilian Portuguese. Therefore, this study performs and describes the translation into Brazilian Portuguese and cross-cultural adaptation of the PQ-16 and the PRIME-Screen.

Method
Prior to beginning the translation process, the authors of the present study obtained consent from the authors of the original screening tools to translate their work. We followed the guidelines proposed by Guillemin et al. 18 for cross-cultural adaptation of psychometric instruments.
The PQ-16 has 16 items -nine items related to perceptual abnormalities, five items related to changes in thought content, and two items related to negative symptoms. 14,19 Respondents mark their response to each item as true or false and also provide an answer on a Likerttype scale that ranks their level of associated distress on a scale from "0" (no distress) to "3" (severe distress).
The PRIME-Screen has 12 items. It is based on the positive symptom domain of the SIPS, which tracks presence of unusual mental activity such as ideas of grandeur and persecution, hallucinatory experiences, and lack of insight. 20 Respondents indicate how much they agree or disagree with the propositions in each item regarding their experiences in the past year using a Likert-type scale ranging from "0" (definitely disagree) to "6" (definitely agree).
Our translation process included the following steps: These steps are illustrated in Figure 1.
As mentioned above, we sent the screening tools to the original authors after back-translating them into English. The original authors did not suggest any changes to the PQ-16 ( Table 1). The original authors did point out some inconsistencies in the PRIME-Screen (Table 2); therefore, we made corrections accordingly, as shown in Tables 1 and 2

Discussion
Previous studies have shown that the PQ and PRIME-Screen are among the best instruments for screening people who are at risk of developing psychosis. 16 19 based on four studies evaluating the diagnostic accuracy of the PQ-16, found that a total symptom score of ≥ 6 and a distress score of ≥ 9 were valid cut-offs for community-based samples. In help-seeking populations, lower scores were suggested for distress