Physiotherapy is a health care profession that addresses the issues of human movement, functionality and quality of life (Melnick 2015). The scope of practice and span of physiotherapy training varies from one country to another (Moffat 2012). For instance, the entry-level qualification in the USA is predominantly a Doctor of Physical Therapy (DPT) (Boissonnault et al. 2014). In Nigeria, the DPT programme has been approved by the National Universities Commission (2018) as the minimum benchmark for entry-level physiotherapy education – a curriculum that deepens the contents of diagnostic imaging training (American Physical Therapy Association [APTA] 2010; Domholdt, Emery & Harris 2004; Medical Rehabilitation Therapists Board of Nigeria [MRTBN] 2009). Musculoskeletal imaging is a subspecialty of diagnostic imaging, which involves ordering and interpreting medical images of bones, joints and associated soft tissues to either identify or rule out pathologies (Reiser, Baur-Melnyk & Glaser 2008).

Musculoskeletal system imaging is indispensable to physiotherapists and has always been a component of their clinical decision-making (Domholdt et al. 2004). It has also been shown that well-trained physiotherapists can utilise diagnostic imaging appropriately as an autonomous provider or within a multidisciplinary team (Kersten et al. 2007; Moore et al. 2005).

The inadequacy in education and training of some referral sources in musculoskeletal disorders and incomplete evaluation of patients for serious pathologies prior to referral for physiotherapy have contributed to the need for physiotherapists to become better at assessment prior to managing patients (Boissonnault et al. 2014).

Physiotherapists’ competency in the utilisation of musculoskeletal imaging for clinical examination purposes was established among the US military physiotherapists 44 years ago (James & Stutart 1975).

Although this level of competency was mostly demonstrated within the military model of practice, examples also exist in civilian settings in the USA and abroad (Moore et al. 2005).

The current practice model in Nigeria is tailored towards autonomous practice, which includes unrestricted direct patient access, the ability to refer to other providers and the ability to refer for diagnostic tests (APTA 2014; MRTBN 2009). However, the Nigerian Nuclear Regulation Authority (2006) warned that diagnostic imaging should be justified by weighing its benefits against potential radiation hazards. Clinicians are advised to always consider available alternatives such as ultrasound and magnetic resonance imaging (MRI), which do not involve medical exposure. Inappropriate utilisation of imaging can lead to over-exposure of patients to radiation hazards and economic wastages (Gazelle et al. 2007; Lehnert & Bree 2010).

Various countries have legislation and regulations over diagnostic imaging (Kam 2005). In the USA, Canada, and some other countries, the Physiotherapy Practice Act has defined the scope of physiotherapy in diagnostic imaging (Boyles et al. 2011; Chong et al. 2015). Physiotherapists in Australia, New Zealand and the United Kingdom have diagnostic imaging referral rights within their roles as primary care providers (Prabhu & Ahmed 2017). Specifically, some states in the USA have an affirmative statement with diagnostic imaging referral right, being part of practice. In other states (such as Oregon and Arizona), radiological investigation is not included in practice acts; in New Jersey and Illinois, it is prohibited, while Colorado proposed disciplinary action against physiotherapists who request X-rays (Boyles et al. 2011). In Nigeria, the Medical Rehabilitation Therapists Act, M9 LFN 2004, does not prohibit physiotherapists from the utilisation of diagnostic imaging in clinical practice.

However, the use of diagnostic imaging by physiotherapists has a sound foundation for expansion in future practice and commensurate emphasis in physiotherapy education (Boyles et al. 2011; James & Stuart 1975).

Accordingly, there is a need to explore the level and nature of training, attitude, competence and utilisation of musculoskeletal imaging among physiotherapists in Nigeria. Following a search of the international literature using keywords to search databases, the authors could not identify any appropriate survey instrument, which could be used for a study such as this. Therefore, a new survey instrument, which can be used to collect comprehensive data on the musculoskeletal imaging profile of physiotherapists in Nigeria, was developed.

An exploratory sequential mixed methods design was undertaken among registered physiotherapists in Nigeria.

A qualitative approach was used to develop the contents of the survey instrument, while a quantitative design was used to validate and pilot test the draft instrument (Dizon, Grimmer-Somers & Kumar 2011).

The first draft of the instrument consisted of the major themes that were raised from the FGD (Table 1). It was organised into six domains and sent to the selected team of experts for validation. The outcome of the validation processes led to the modification in chronology, the addition of new questions and expansion of some response options. For instance, the panellists suggested that ‘not applicable’ should be included among the answer options for the items in part B. This was because some of the questions were found to be not applicable to all the potential respondents.

The face and content validated instrument was divided into (parts A–F) six domains as follows: (A) 17 questions on demographic details which include age, gender, marital status, licence renewal history, years of practice, the region of practice, practice setting, employment cadre, the specialty of interest, continuous professional development and educational background; (B) 25 questions on nature of training in musculoskeletal imaging which includes entry points, methods, duration, personnel employed and hands-on experiences for undergraduates, internship, workshop and postgraduate training; (C) seven questions on the level of training in the interpretation of musculoskeletal imaging results; (D) eight questions on the attitude of physiotherapists towards musculoskeletal imaging; (E) seven questions on the utilisation of musculoskeletal imaging results in clinical practice and (F) six questions on the level of competence of physiotherapists in the interpretation of musculoskeletal imaging results.

In total, the questionnaire was made up of 70 items. Parts A and B are multiple choices or dichotomous questions (yes or no). This aspect was designed for descriptive purposes and did not have a specific scoring system. The remaining domains (parts C–F) were designed as a five-point Likert scale (1 – lowest to 5 – highest score). Specifically, the range of score for the domains was: part C (6–30), part D (8–40), part E (7–35) and part F (6–30).

Another essential component of the methodology was piloting the PMIPQ on representatives of the study population (Dizon et al. 2011). Thirty-one physiotherapists (12 women and 19 men) averagely aged 33.81 ± 5.04 years and practice experience of 7.58 ± 5.16 years participated in the pilot survey. The average time taken to complete the online version of PMIPQ was 9.56 ± 1.51 min. The amount of time taken for completing the hard copy version of the instrument could not be analysed because it was co-administered with a criterion and a short checklist (Table 2) – to critique the general properties of the questionnaire.

Information from the checklists showed that a few respondents requested a revision of the language structure of part B for clarity. Nonetheless, the summary of the checklist (Table 2) showed that the PMIPQ was relevant, simple, all-inclusive, chronological, self-explanatory and easy to complete. The instrument has good psychometric properties: face, content and criterion validity (Table 3). The internal consistency (a-score) and the test–retest reliability (ρ-score) across the domains were: part C (α = 0.731, ρ = 0.973), part D (α = 0.737, ρ = 0.979), part E (α = 0.446, ρ = 0.842) and part F (α = 0.796, ρ = 0.716).

Our study explored the current musculoskeletal imaging practice among physiotherapists in Nigeria. We could not find any instrument in the literature that could address the objectives of our proposed study. Therefore, the development of the PMIPQ was undertaken. The questionnaire was designed, validated and piloted in adherence with the general guidelines for development and implementation of (face-to-face and online) dual administration mode surveys (Adje 2016; Andrews et al. 2003; Dizon et al. 2011; Potter et al. 2012). The PMIPQ was developed to collect data on the nature of training, attitude, competence and the level of utilisation of musculoskeletal imaging among physiotherapists in Nigeria and countries with similar education and practice models. This instrument is relevant in light of the growing influence of musculoskeletal system imaging in physiotherapy practice worldwide (Boissonnault et al. 2014).

The scope of practice and span of physiotherapy training vary from one country to another (Moffat 2012). Similarly, the curricula and diagnostic imaging contents vary among training institutions within and across countries (Boissonnault et al. 2014). Physiotherapy entry-level qualification in some countries does not involve internship training (Onigbinde 2006). In Nigeria, the post-entry-level internship programme is mandatory for all physiotherapists since its commencement in 1994. Therefore, the information on the nature of musculoskeletal imaging training during the internship was not applicable to respondents who had graduated earlier. Accordingly, the result of the pilot survey led to the adjustment of PMIPQ to accommodate respondents who did not have internship training.

All the domains (parts) of the online version of the instrument were programmed as compulsory fields except for part B where a respondent can choose the ‘not applicable’ option and move on to the next subsection. The software ensures that incomplete questionnaires cannot be submitted; rather, the missing fields will be highlighted for the respondent to complete them before submission. There is no question that this innovation reduces the problem of incomplete data – a major challenge in questionnaire-based surveys (Andrews et al. 2003). The PMIPQ was reported to be self-explanatory, easy to administer and appealing to respondents (Table 2). Boynton (2004) has stated that poor understanding of a questionnaire can lead to low response rates or incomplete data.

The results of our study showed a positive correlation between the criterion and all the tested domains of the PMIPQ. However, part E (ρ = 0.515, p = 0.006) and F (ρ = 0.481, p = 0.039) have a statistically significant and stronger positive correlation with the criterion. This outcome was expected because the criterion has a closer construct with parts E and F. The criterion was specifically designed to obtain opinions on competence and level of utilisation (of imaging modalities among physiotherapists) which corresponds with the objectives of parts E and F of the PMIPQ, respectively.

Nonetheless, all the domains of the online version of the PMIPQ showed an acceptable level of internal consistency except part E, which obtained responses on the level of utilisation of imaging studies. The value α = 0.446 fell below the cut-off value of α = 0.5. This can be explained by the fact that there is no standard practice act regarding the utilisation of imaging among physiotherapists in Nigeria. Instead, hospitals have diverse internal policies on the scope of practice. However, there was a high reliability score across all the domains (Table 3). Reliability was obtained by a pair-wise correlation between the respondents’ scores on the paper and the online versions of the PMIPQ.

The emphasis PMIPQ laid on ultrasonography as content in physiotherapy training and practice is noteworthy.

This is because ultrasound is a promising area for future practice. It has the comparative advantage of being portable, inexpensive and safe, as well as being a non-ionising-radiation-based musculoskeletal imaging modality (Boyles et al. 2011; Chong et al. 2015; Potter et al. 2012).

This study combined a qualitative and quantitative approach in analysing the outcome of the pilot testing; a similar study that reported the processes of development of a new instrument relied only on a qualitative approach (Dizon et al. 2011). Data entry and coding are recommended components of pilot testing as this allows troubleshooting of possible problems in data management and analysis (Boynton 2004). Hence, this methodology appears unique and may interest other researchers who may wish to carry out a similar study.

This study has provided a relevant instrument for assessing the musculoskeletal imaging profile of physiotherapist in Nigeria and abroad. The PMIPQ has acceptable psychometric properties, the design is comprehensive and mode of administration is innovative and appealing to respondents.