How can work motivation among primary care physicians be assessed? Development and psychometric validation of the M-PCP scale: a cross-sectional study in the Girona Health Region, Spain

Abstract

Background

Work motivation is essential for well-being, performance and retention in general practice. Despite its relevance, no validated instrument is available to assess motivation specifically among primary care physicians (PCPs) or to inform targeted managerial action.

Aim

To develop and validate the Motivation Scale for PCP (M-PCP), a theory-driven instrument designed to capture motivational processes relevant to general practice.

Methods

The scale was developed using a structured, theory-based process grounded in Ford’s Motivational Systems Theory. Items were generated from existing models, reviewed by experts and tested in a cross-sectional survey of PCPs in the Girona Health Region (Catalonia, Spain). Psychometric analyses included item discrimination, exploratory factor analysis and partial least squares structural equation modelling.

Results

After removing items with low discrimination or weak loadings, the final 37-item scale showed a clear two-factor structure: goals (continuous improvement, self-care and agency) and perceptions (emotions and contextual agency beliefs). Both constructs showed strong reliability and validity. The model fit was acceptable, and perceptions explained a moderate proportion of variance in goals. These findings indicate that the scale captures key motivational processes linked to daily clinical work.

Conclusions

The M-PCP is a valid tool for assessing work motivation in general practice. By identifying goal-related, emotional and contextual motivational processes, it supports targeted strategies that may enhance clinical performance, continuity of care and physician well-being. Further research should examine its predictive value and applicability across health systems.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• The Motivation Scale for Primary Care Physicians fills an important gap by measuring work motivation in primary care as a multidimensional construct.

• It is theory-driven and captures goals, emotions and contextual agency beliefs, strengthening its relevance for general practice.

• The scale has practical value for managers because it can help identify motivational needs and support targeted interventions to improve workplace motivation and reduce turnover.

• Developed and validated in one European region, so its generalisability to other health systems and cultural settings remains to be confirmed.

• Further longitudinal and external validation studies are needed to test predictive validity and stability over time.

Background

Work motivation is essential for professional well-being, performance and the sustainability of general practice. In contexts where continuity, accessibility and person-centred care are core principles, maintaining physician motivation is key for care quality and stable teams.^{1 2} Across Europe, rising workload, administrative burden and persistent shortages of general practitioners have intensified concerns about workforce stability.^{3 4} Global trends show increasing burnout, emotional exhaustion and job dissatisfaction among primary care physicians (PCP), driven by time pressure, limited career progression and high emotional demands,5,8 contributing to absenteeism, turnover intentions and enduring vacancies.9,11

European system assessments warn that workforce shortages increasingly threaten the sustainability, quality and safety of general practice.^{3 4 12} National analyses, including the 2024 report from the Spanish Society of Public Health, also highlight the need to strengthen working conditions and prevent attrition.¹³ Consistently, the WHO’s Global Strategy on Human Resources for Health: Workforce 2030 stresses that motivation and retention are essential for universal health coverage.¹⁴

Although many interventions to improve physician well-being focus on individual strategies—such as coping skills, resilience training or stress-management programmes—their impact on long-term retention and performance is limited because they do not address the systemic conditions that drive demotivation.^{6 8 15 16} Contemporary frameworks, including Franco et al’s model of health worker motivation and the National Academy of Medicine’s systems approach, emphasise that motivation arises from interactions among individual, organisational and policy-level determinants.^{17 18} Within this perspective, management practices, organisational climate and resource availability become central influences on motivation, retention and professional commitment in general practice.

Existing instruments—such as the Work Extrinsic and Intrinsic Motivation Scale (WEIMS),¹⁹ Multidimensional Work Motivation Scale (MWMS),²⁰ Copenhagen Psychosocial Questionnaire (COPSOQ)²¹ and engagement scales like the Utrecht Work Engagement Scale (UWES)²²—provide useful information on psychological orientation, psychosocial risks or engagement. However, they present important limitations for general practice: they are not tailored to the clinical and organisational realities of primary care,^{1 2} they do not assess motivation as a multidimensional psychological and organisational construct,^{23 24} and they offer limited applicability for informing retention efforts or prioritising practical actions within primary care teams.25,27 This gap highlights the need for a context-specific, theoretically grounded instrument capable of capturing the motivational processes relevant to general practitioners.

Evidence from low- and middle-income countries also shows that health worker motivation is shaped by working conditions, organisational support and perceived career prospects. Studies conducted in settings such as Jordan, Georgia,²⁸ Kenya²⁹ and Pakistan³⁰ have described the importance of both intrinsic and extrinsic determinants of motivation, as well as their consequences for performance, retention and service quality. These findings reinforce the view that motivation is a context-dependent construct and support the need for instruments that capture not only psychological orientation, but also the organisational and relational factors that influence day-to-day work.

The Motivation Scale for PCP (M-PCP) was developed to address this gap. Grounded in motivational science—particularly Ford’s Motivational Systems Theory²⁵ and informed by evidence on health worker motivation,^{17 23 31 32} the M-PCP assesses motivational goals, emotional experiences and contextual agency beliefs specific to general practice. The objectives of this study were: (a) to design the M-PCP through a theory-driven and contextually relevant process; (b) to validate its psychometric properties and examine convergent and divergent validity and (c) to offer a context-specific instrument that helps general practice teams better understand motivational dynamics and identify areas for improvement in daily work.

Methods

Study design

This study followed a cross-sectional instrument-development design with two consecutive phases:

• Phase 1: Scale development.

  An initial item pool was generated de novo through a structured, theory-driven process based on Ford’s Motivational Systems Theory, Self-Determination Theory, previous studies,³³ and models of health-worker motivation. The items were then reviewed by experts to ensure content validity, clarity, cultural relevance and applicability to PCPs. All items were newly developed by the author specifically for this study and were not directly adapted from or reproduced from previously published instruments.

• Phase 2: Psychometric validation.

  A formal pilot study was not conducted; the expert validation phase served as a preliminary testing stage, allowing refinement of item clarity and relevance prior to field administration. The revised version was then administered to PCPs in the Girona Health Region. Reliability, factorial structure and convergent/divergent validity were assessed using exploratory factor analysis and partial least squares-structural equation modelling (PLS-SEM), supported by complementary instruments.

This sequential structure enabled the integration of conceptual development with quantitative validation, producing a context-specific and psychometrically robust instrument.

Participants

The target population included all PCPs working in the public healthcare network of the Girona Health Region (Catalonia, Spain). A census sampling strategy was used: a single survey link was distributed electronically through institutional channels, allowing all physicians to participate. Although a census sampling strategy was used, representativeness cannot be fully ensured due to voluntary participation. However, the sample included physicians across different contract types and experience levels, reflecting the structural composition of the primary care workforce in the region. This supports a reasonable approximation to the target population, although selection bias cannot be excluded.

A total of 112 responses were received. After excluding eight respondents who declined consent and two who submitted incomplete questionnaires, 102 valid questionnaires were retained. Participation was voluntary, anonymous and uncompensated. Sociodemographic variables included sex, years of experience and contract type (permanent or temporary).

Although all PCPs in the Girona Health Region were invited, the exact number of active PCPs in the public system is not publicly available, preventing calculation of a precise response rate. The final sample size of 102 participants was adequate for the planned psychometric analyses and aligns with international recommendations for initial validation studies. These include COSMIN methodology, which considers samples of at least 100 participants adequate for the initial validation of measurement instruments,³⁴ PLS-SEM criteria,³⁵ which emphasise statistical power rather than item count. Indicating that for using standard power analysis for PLS-SEM, models with small-to-medium effects require samples of around 90–110 participants, which aligns with the sample and classical psychometric literature which classifies samples of around 100 as ‘good’ for exploratory or confirmatory factor analysis (CFA)³⁶ and as acceptable for theory-driven models with strong loadings and moderate complexity.³⁷ This supports the adequacy of the sample for psychometric purposes. However, because participation was voluntary and the denominator was unknown, the representativeness of the samples cannot be formally established. A flow diagram describing the selection of participants is presented in figure 1.

Figure 1. Flow diagram. PCPs, primary care physicians.

Instrument development

The M-PCP was developed through a structured, theory-driven process. The conceptual framework was primarily based on Ford’s Motivational Systems Theory,²⁵ complemented by evidence from previous models of health worker motivation^{17 23} and organisational psychology.^{38 39} The item pool was generated de novo by the author specifically for this study and was designed to capture three core components: (a) goals (continuous improvement, self-care and systemic agency), (b) emotions related to professional activity and (c) personal agency beliefs regarding contextual and organisational conditions.

Item generation involved drafting an initial pool of items for each variable based on the theoretical framework from a previous study,³³ which defined the main motivational dimensions and indicators. Wording was adapted to ensure clarity, contextual relevance and applicability across the full range of PCP activities.

An expert panel was convened to assess content validity. Eight experts were purposively selected based on predefined criteria: professional experience in at least one of the following fields—work motivation, research methodology, human resources management in healthcare or psychometric test validation.^{40 41} The number of experts was decided pragmatically and is consistent with established methodological recommendations for content validation studies, which commonly use small expert panels, often ranging from 5 to 10 experts. This approach follows established methodological recommendations for instrument development and content validation.^{34 41} The scale was submitted for content validation using the methodology proposed by Escobar-Pérez and Cuervo-Martínez,⁴⁰ five completed the assessment.

Their feedback led to the elimination of redundant items, refinement of wording and the inclusion of explanatory statements in the goals subscale. These modifications strengthened the face and content validity of the instrument prior to field testing. Online supplemental appendix A presents the initial item pool of the scale prior to revisions made following expert content validation. All items were newly developed for this study and do not reproduce previously published instruments.

The resulting version of the M-PCP was structured into dimensions and variables, summarised in table 1, which includes definitions, abbreviations and the initial number of items per variable. A 10-point Likert-type response scale was deliberately chosen to increase sensitivity, allowing detection of subtle variations in motivational perceptions and providing greater precision in measuring motivational experiences.

Table 1. Definitions for the different dimensions of the M-PCP scale.

Dimension	Variable (abbrev.)	Definition	Initial items
Goals	Continuous improvement (MC)	Participation in activities aimed at improving patient care quality. Includes training, evaluation of health indicators and prosocial motivation to protect patient well-being.1 28 38 48	9
	Self-care and participation in projects (SC)	Voluntary involvement in innovation, teaching, community activities and practices that promote physician well-being.32 38 49 50	7
	Systemic agency (AG)	Contributions aimed at transforming practices at individual, organisational or social level, reflecting perceived capacity to influence the system.38 51	7
Emotions	Affect (EM)	Emotional impact of working conditions (positive/negative), linked to stress, satisfaction and engagement.32 38 50	12
Personal agency beliefs	Context beliefs (PC)	Perceptions of support or barriers in the environment when pursuing goals.1 38 51	15
	Capability beliefs	Indirectly assessed through effort and persistence; aligned with Bandura’s self-efficacy theory.38 51	Not directly operationalised

AG, Systemic agency; EM, emotions; MC, Continuos improvement; M-PCP, Motivation Scale for Primary Care Physicians; PC, profesional context ; SC, Self-care and participation in projects.

Complementary instruments

To assess the convergent and divergent validity of the M-PCP scale, four validated instruments were included:

• UWES-9²²: Due to a data entry error, six items were used, still covering vigour, dedication and absorption. Engagement was considered a convergent measure of positive motivation.

• MWMS²⁰: Only the extrinsic regulation subscale was applied, assessing work driven by external pressures or rewards. It was used as a divergent measure.

• COPSOQ-ISTAS21, Spanish version⁴²: Six items assessed psychosocial risks and job satisfaction in primary care, situating motivation within its broader work context.

• Intention to quit⁴³: Two items adapted from D’Alleva Chiorri et al measured intention to leave the job or profession, included as a behavioural outcome of motivation.

Together, these instruments provided complementary evidence for evaluating the convergent, divergent and criterion-related validity of the M-PCP scale.

Data collection

Data were collected using the REDCap platform,⁴⁴ provided by the Hospital de Palamós/Serveis de Salut Integrats Baix Empordà (SSIBE). A single survey link was generated and distributed in November 2024 via institutional email to the heads of Basic Health Areas in the Girona Health Region, who subsequently forwarded it to PCPs.

The survey remained open for 8 weeks, and two reminders were issued during this period, the full questionnaire can be found in online supplemental appendix B. Participation was voluntary and anonymous. Before accessing the questionnaire, participants received an information sheet describing the study objectives and were required to provide electronic informed consent (online supplemental appendix C). Only individuals who agreed to the statement “I have read the information sheet and agree to participate in the study” were able to proceed.

In addition to the M-PCP scale and complementary instruments, the survey collected basic sociodemographic variables, including sex, years of professional experience and type of employment contract (permanent or temporary).

Data analysis

Data were exported from REDCap into SPSS V.29 for initial cleaning, including checks for out-of-range values, recoding of negatively worded items and analysis of missing data. All M-PCP items were mandatory; missing values in complementary instruments were handled using multiple imputation procedures.⁴⁵

Descriptive statistics (mean, SD, variance, skewness and kurtosis) and item discrimination indices were computed in JASP V.0.18. Items with discrimination values <0.30 were removed.⁴⁶ Internal consistency was assessed using Cronbach’s alpha and McDonald’s omega.^{47 48} Shapiro-Wilk tests indicated non-normal distribution across M-PCP variables.

Exploratory factor analysis (EFA) was conducted using principal axis factoring (PAF) with Promax rotation (factor loadings ≥0.40), supported by the Kaiser-Meyer-Olkin statistic and Bartlett’s test of sphericity.⁴⁹ The EFA indicated two main factors: (1) goals and (2) perceptions of the professional context and affective impact.

Subsequently, CFA was performed using PLS-SEM in SmartPLS V.4.0, appropriate for models with non-normal data and modest sample sizes.³⁵ Latent variables were reflective and organised into second-order constructs: (a) goals (continuous improvement, self-care, systemic agency), (b) emotions and (c) perceptions (professional context). A two-stage approach was used to estimate the hierarchical structure.

Missing data occurred only in the complementary instruments and represented <0.5% of all values. A missing-data pattern analysis was conducted to explore the magnitude and distribution of the missing values. The results showed a very low proportion of missing data, and the missing values appeared in an isolated, scattered manner, without clustering within the same cases or affecting multiple variables simultaneously. This pattern suggested the absence of a structured mechanism linked to a specific subgroup of participants and was considered compatible with the missing at random assumption. Multiple imputation was then applied to the complementary instruments before conducting the validity analyses.

Convergent and divergent validity were examined through Spearman correlations between M-PCP subscales and complementary instruments (UWES, MWMS, COPSOQ and intention to quit). Additional descriptive analyses were conducted for job satisfaction and intention to quit.

Ethics approval

The study was conducted in accordance with the ethical principles of the Declaration of Helsinki. The research protocol was first reviewed by the Research Committee of the Hospital de Palamós/Serveis de Salut Integrats Baix Empordà (SSIBE, code 55923/23) and subsequently by the Ethics Committee for Research with Medicines of Girona (CEIm-Girona, code 2023.228). Because the study was developed in multiple phases, both committees required a second review of the final version of the scale prior to its distribution, leading to two separate approvals.

Participation was voluntary and anonymous. Electronic informed consent was obtained before participants accessed the questionnaire, and progression through the survey was not possible without explicit agreement. All data were handled confidentially and in full compliance with the General Data Protection Regulation (EU 2016/679). Questionnaire data were stored securely in the REDCap platform, with technical access restricted exclusively to the authorised members of the research team. No identifiable personal data were collected at any stage.

Patient and public involvement

The instrument was applied in PCPs working in the province of Girona in the autonomous community of Catalonia in Spain.

Findings

Psychometric analysis

Descriptive analysis

An initial descriptive analysis was conducted to evaluate item performance. Discrimination indices and coefficients were calculated for all items, and those scoring below the established 0.30 threshold were removed from the scale. A total of 11 items that failed to meet the discrimination threshold (corrected item-total correlation <0.30). Specifically, these were items from the continuous improvement (4), emotions (6) and professional context (1) subscales (see online supplemental table 1).

Reliability analysis

After item reduction, a new descriptive and reliability analysis was performed (table 2). The final version of the M-PCP showed high reliability, with a Cronbach’s alpha of α=0.932 and an Omega coefficient of ω=0.931 for the full scale. Subscale reliability estimates were also high: Goals dimension: α=0.884, ω=0.888, Continuous Improvement: α=0.702, ω=0.698, Self-care: α=0.789, ω=0.796, Agency: α=0.837, ω=0.839, Emotions: α=0.827, ω=0.838, Professional Context: α=0.886, ω=0.889. All values fall within the recommended range (0.70–0.95), indicating strong internal consistency without redundancy, according to Streiner’s reliability criteria.⁴⁸ Item-rest correlations were satisfactory across subscales, and no item deletion would have substantially improved reliability. Full item-level descriptive statistics (means, variances, kurtosis, item–rest correlations, α and ω if item deleted) are available in online supplemental table 2.

Table 2. Demographic data.

Gender	Seniority	N	%
Women	<5 years	0	0
	5–10 years	13	17.1
	>10 years	63	82.9
Men	<5 years	3	11.5
	5–10 years	3	11.5
	>10 years	20	76.9

Exploratory factor analysis

Sampling adequacy for factor analysis was confirmed: the Kaiser-Meyer-Olkin (KMO)⁴⁹ value was 0.788, indicating a ‘good’ level of common variance, and Bartlett’s test of sphericity was significant (χ²=2306.57, p<0.001), supporting factorability. No individual MSA values fell below the 0.50 threshold.

An EFA was conducted using PAF with Promax rotation and a loading criterion of >0.40 using JASP V.0.18. A two-factor solution emerged, explaining the underlying structure coherently:

• Factor 1 (‘perceptions’) grouped items from the professional context and emotions dimensions, reflecting perceived organisational conditions and their affective impact.

• Factor 2 (‘goals’) included all items related to continuous improvement, self-care/participation in projects and agency.

All retained items loaded above the 0.40 threshold on their respective factors, with no cross-loadings of concern. The EFA results were consistent with the theoretical model and justified proceeding to confirmatory analysis. Item loadings are reported in online supplemental table 3.

SEM analysis

PLS-SEM was performed using SmartPLS V.4.0 to validate the hierarchical structure of the M-PCP. Given the presence of second-order constructs, a two-stage approach was applied following recommended guidelines.³⁵ Analyses proceeded in two steps: assessment of the measurement model and evaluation of the structural model.

Measurement model quality was first assessed. Two additional items (V51PC and V53PC) were removed due to weak factor loadings (<0.40), which compromised the convergent validity of their respective constructs.³⁵ Multicollinearity was acceptable, with all item-level variance inflation factor (VIF) values below 5 (online supplemental figure 1). Internal consistency was strong, with Cronbach’s alpha coefficients of 0.812 for GOALS and 0.872 for PERC, and composite reliability values (RhoC) of 0.887 and 0.939, respectively. Convergent validity was supported by Average Variance Extracted (AVE) scores exceeding .50 (GOALS=0.725; PERC=0.886).

All first-order constructs (continuous improvement, self-care, agency, emotions and professional context) showed external loadings above 0.70, and their composite reliability values ranged from 0.813 to 0.898. Discriminant validity was confirmed using the Fornell-Larcker criterion, as the square root of the AVE for GOALS (0.851) and PERC (0.941) exceeded their inter-construct correlation (0.543). The HTMT value (0.628) was below the 0.85 threshold, further supporting discriminant validity. Construct-level VIF values were also acceptable (all <3). Figure 2 displays the first-order model.

Figure 2. First-order measurement model. AG, agency; EM, emotions; MC, continuous improvement; PC, professional context; PERC, perceptions; SC, self-care.

Once the measurement model was established, the structural model was evaluated. The standardised root mean square residual was 0.079 for both saturated and estimated models, within the acceptable <0.08 range. Model fit was additionally supported by a geodesic distance (dG) of 0.089. The model explained 29.5% of the variance in GOALS (R²=0.295; adjusted R²=0.288), representing a moderate explanatory capacity with minimal evidence of overfitting. The effect of PERC on GOALS showed an F² of 0.418, which is considered large (>0.35 per Cohen⁵⁰). Given the cross-sectional nature of the data, all relationships should be interpreted as associative rather than causal. The fit of the structural model was evaluated in order to verify the relationship between the constructs (figure 3).

Figure 3. Second-order structural model of de M-PCP. AG, agency; EM, emotions; MC, continuous improvement; M-PCP, Motivation Scale for Primary Care Physicians; PC, professional context; PERC, perceptions; SC, self-care.

Discriminant validity analysis of the M-PCP scale

Discriminant validity was examined through Spearman’s correlations between the M-PCP subscales and external measures, including engagement, external regulation, mental health, job satisfaction and intention to quit (figure 4). Correlations among the M-PCP subscales were moderate to high, supporting convergent validity. As expected, all subscales showed significant negative correlations with intention to quit, consistent with divergent validity.

Figure 4. Heat map of Spearman correlations between the variables evaluated in the study. *p<0.05, **p<0.01, ***p<0.001. AG, agency; EG, engagement; EM, emotions; ER, external regulation; MC, continuous improvement; MH, mental health; PC, professional context; QUIT, intention to leave; SAT, job satisfaction; SC, self-care.

Associations with external regulation and mental health were low or nonsignificant, further supporting discriminant validity. The emotions (EM) and professional context (PC) subscales correlated moderately with engagement (r=0.61 and r=0.43, respectively). The highest inter-subscale correlation within the M-PCP was observed between EM and PC (r=0.727), a value that—although close to the recommended caution zone (> 0.70)—remains below the COSMIN³⁴ threshold of 0.85 for construct distinctiveness. This indicates that emotional experience and contextual perceptions are related but not conceptually redundant. All remaining correlations were below thresholds associated with multicollinearity, supporting adequate discriminant validity across constructs.

Descriptive information

Of all respondents, 74.51% identified as women, 81.37% had more than 10 years of professional experience and 89.22% held a permanent position. Table 2 shows the demographic data.

Regarding job satisfaction, 26.47% reported feeling dissatisfied or very dissatisfied with their work. In relation to turnover intentions, 53.47% of participants reported that they had considered leaving their current job on occasion, while 3.96% stated that they intend to leave within the next 2 years. When asked about leaving the medical profession entirely, 37.25% indicated that they had considered this possibility on occasion.

Discussion

Principal findings

This study developed and validated the M-PCP, a theory-driven instrument assessing motivational processes specific to general practice. Psychometric analyses supported a robust 37-item structure organised into two higher-order constructs—Goals and Perceptions—with high internal consistency, satisfactory factorial validity, and adequate convergent and divergent validity. Although the explained variance was moderate, the model identified a meaningful and statistically strong relationship between perceptions and goals, supporting the theoretical coherence of the scale. The final version of the M-PCP scale is provided in online supplemental appendix D, including all items and the response format. The instrument is original, was developed specifically for this study and is available for research use with appropriate citation.

The M-PCP captures motivational dynamics that underpin clinical performance, continuity of care, relational quality and engagement in improvement activities, all core priorities in general practice. Prior evidence shows that physicians’ well-being and motivational states are closely linked to patient safety, quality of care and clinical effectiveness.^{51 52}

Beyond its measurement function, the M-PCP provides insight into workload, emotional strain, professional development opportunities and contextual support—organisational determinants known to shape care quality in primary care. By identifying these patterns, the scale aligns with contemporary calls for integrated models connecting professional motivation with clinical outcomes and the functioning of primary care teams.

Comparison with previous literature

Existing instruments—such as WEIMS,¹⁹ MWMS,²⁰ (COPSOQ²¹ and the UWES²²)—are widely used in organisational and health research. However, they primarily assess psychological orientations, psychosocial risks or engagement, offering limited capacity to capture the systemic and practice-specific determinants of motivation relevant to general practice.

In contrast, the M-PCP is explicitly tailored to primary care, integrating individual goals, emotional experiences and contextual agency beliefs that reflect the realities of clinical work. This multidimensional structure aligns with international calls for more comprehensive models of health-worker motivation^{17 23} and is grounded in Ford’s Motivational Systems Theory,²⁵ which conceptualises motivation as the interaction between goals, emotions and personal agency within specific environments.

By incorporating these dimensions, the M-PCP advances the field: rather than serving as a generic motivational measure, it links motivational assessment with workforce sustainability, professional performance and organisational conditions that are central to effective primary care systems.

Implications for research and practice

The M-PCP scale provides managers with a structured way to identify motivational strengths and vulnerabilities among general practitioners. By differentiating between goals, emotional experiences and contextual agency beliefs, the instrument supports targeted interventions rather than universal approaches, which have shown limited and inconsistent effectiveness in previous well-being and motivation programmes for PCPs.⁵³

From a practical perspective, the M-PCP scale can be used as a diagnostic tool to identify specific motivational deficits within clinical teams. Its ability to generate granular profiles facilitates a shift from generalised, one-size-fits-all interventions to more precise managerial responses. By showing whether reduced motivation arises primarily from personal goals or contextual perceptions, the scale helps organisations design more effective and tailored support strategies.

As detailed in the strategic roadmap provided in table 3, these indicators guide specific managerial actions. For instance, a team exhibiting low 'Contextual Agency’ would benefit more from structural changes in workload and resource allocation than from individual resilience training. Conversely, low scores in ‘self-care’ objectives should trigger organisational support for professional development and well-being programmes. Ultimately, the M-PCP scale enables managers to prioritise interventions that directly address the most salient motivational needs identified in their medical staff.

Table 3. Organisational applications of the M-PCP scale and recommended managerial responses.

Dimension	M-PCP indicators	Target group identified by the scale	Recommended managerial strategies	Expected outcomes
Goals	Self-care	PCP with low scores in involvement in self-care activities or teaching/community projects.	Facilitate access to structured self-care and well-being programmes; promote participation in teaching, research and community projects; include self-care objectives in performance appraisal.	Improved work–life balance, reduced risk of burnout and a greater sense of purpose and belonging to the team and the community.
	Agency	PCP who indicate low influence within the organisation or limited participation in change processes.	Create formal spaces for participation in committees, improvement groups and decision-making forums; provide training in clinical leadership and change management; give visible follow-up to their proposals.	Increased sense of influence and control over the system, greater commitment to reforms, and greater willingness to promote organisational innovations.
Perceptions	Emotions	PCP with high levels of stress, fatigue or negative affect.	Implement stress-management and resilience programmes, mindfulness training and peer-support groups (eg, Balint groups); improve feedback and institutional recognition.	Reduced burnout and emotional fatigue, increased resilience and an improved emotional climate within the team.
	Contextual beliefs	PCP who perceive low autonomy, high administrative burden or lack of resources.	Promote participatory decision-making; redistribute administrative tasks; improve resource allocation and organisational support channels; review clinical workload and patient lists.	Greater job satisfaction and perceived fairness; reduced intention to leave, turnover and absenteeism.

M-PCP, Motivation Scale for Primary Care Physicians.

From a research perspective, the scale offers a theoretically grounded measure for studying motivational processes in primary care and supports comparative analyses across regions or health systems. Its multidimensional structure enables examination of how specific motivational domains relate to organisational outcomes such as job satisfaction, intention to quit or engagement. Future studies may replicate validation in other contexts, explore longitudinal motivational trajectories and test targeted interventions informed by M-PCP profiles.

Strengths and limitations

This study benefits from a rigorous, theory-driven development process, expert review and robust psychometric validation using complementary instruments and combined EFA–PLS-SEM analyses. Census-based recruitment helped reduce sampling bias. Nevertheless, an important limitation is that the representativeness of the sample cannot be formally established. Participation was voluntary, and the exact denominator of active PCPs in the public system at the time of the survey was not available. Consequently, self-selection bias cannot be ruled out, as physicians with greater interest in motivation may have been more likely to participate. Although the sample size was psychometrically adequate for validating the M-PCP scale, the findings should be generalised to other regions or health systems with caution until further external validation is conducted.

A limitation of this study is the use of a truncated six-item version of the UWES-9 due to a technical error during the survey design phase. While the items included represent the three core dimensions of engagement—vigour, dedication and absorption—and the observed correlations were statistically significant and in the expected direction, this truncation may reduce the robustness of the convergent validity analysis. Future studies should replicate these findings using the full UWES-9 scales to further consolidate the relationship between motivation and engagement among PCPs.

However, the sample was limited to one region, which restricts generalisability and requires external validation in broader settings. The cross-sectional design precludes causal inference or assessment of temporal stability, and future longitudinal studies are needed to examine predictive validity. Additional cognitive testing may improve interpretability in other contexts.

Conclusions

The M-PCP scale demonstrated strong reliability and validity, offering a theory-based instrument for assessing core motivational processes in PCPs—goals, emotional experiences and contextual agency beliefs. Beyond measurement, it can guide targeted organisational strategies to strengthen motivation, well-being and retention. Although validated in one region, its conceptual foundations support application in other health systems. Future studies should evaluate its predictive value, longitudinal stability and cross-cultural performance to consolidate its role in primary care workforce development.

Data availability statement

Data are available on reasonable request. Data may be obtained from a third party and are not publicly available. All data relevant to the study are included in the article or uploaded as supplementary information.