Individuals who purchased medicines at chain pharmacies located in nine districts of Can Tho City.

Customers who have bought drugs at least once from pharmacies of pharmaceutical retail chains. Respondents must have practical experience with and an interest in the chain’s products and services (Moreira and Silva 2015).

Exclude responses that are incomplete, invalid, left blank, or not filled in proper sequence. Also exclude respondents who have not actively sought information in recent years.

We selected the “theory of planned behavior (TPB)” as our theoretical foundation. This theory, founded and developed by social psychologist Ajzen, holds significance in the field of mentality and behavioral science” (Ajzen 1991). TPB is widely used in research aimed at predicting and explaining human behaviors (Zhang and Zhou 2019). Besides subjective norms and behavioral control, another factor that can influence patient loyalty is trust. The relationship marketing literature has revealed that trust is a key factor in building customer loyalty (Larsson and Wilde‐Larsson 2010; Vinagre and Neves 2010) (Fig. 1).

Figure 1. 

Research model.

Hypothesis: Product quality, brand name, price, facilities, convenience, advertising, knowledge of employees, and salesmanship have a positive impact on customer loyalty toward pharmacy chains in Can Tho City.

This study utilized a cross-sectional descriptive research design and collected data through direct phone interviews, Google Forms, or printed questionnaires with clients who purchased drugs at pharmacy chains or hospitals in Can Tho City. The interviews were conducted using a set of questions to evaluate loyalty and factors influencing clients’ decisions to return to the pharmacy chain for future purchases. The survey questionnaire is designed based on outcome-oriented questions. Some previous studies included additional clarifying questions such as, does subjective norm affect customer loyalty? Does perceived behavioral control affect customer loyalty? Does trust affect customer loyalty? Or identify some key factors that make the difference between service quality and satisfaction (Vinagre and Neves 2010; Abrahamsen Grøndahl et al. 2013; Sumaedi et al. 2015).

The first part of the questionnaire is a closed-ended type, where most questions require respondents to choose from predefined options, typically using checkboxes (✓). It focuses on collecting demographic information such as full name, gender, year of birth, area of residence, occupation, and income level. Some questions permit limited open-ended responses; for instance, the income question includes an “Other (please specify)” option, which gives participants room to elaborate if the provided choices are not applicable. This structured design allows for standardized data collection, making analysis and interpretation more efficient. To ensure validity, responses should be checked for completeness, consistency, and relevance (e.g., avoiding blank fields or conflicting answers). Clear instructions and proper quality control during data collection can help improve the reliability of the survey results and identify statistically significant factors that have an impact on the scale/dependent variable. Respondents were required to fill in the available spaces with their information.

The second part involves assessing a combination of factors affecting customer loyalty in chain systems. These factors include product quality (3 items), pharmacy chain brand (3 items), price (3 items), facilities (3 items), convenience (3 items), promotional programs (3 items), along with the professional knowledge of the staff (3 items), and salesmanship (4 items) (Grøndahl et al. 2011). To evaluate these factors, a relatively popular measurement scale in scientific research questionnaires was used to verify individual opinions, behaviors, and perceptions. This section of the questionnaire uses a Likert scale-based instrument designed to measure respondents’ level of agreement with specific statements related to product quality and brand trust. Each item provides five response options: 1 = “strongly disagree,” 2 = “disagree,” 3 = “neutral,” 4 = “agree,” and 5 = “completely agree” (Kondasani and Panda 2015). This format supports the validation of customer experience by providing consistent response categories, and it enables researchers to measure sentiment or satisfaction levels in a standardized way.

The scale achieved reliability with a Cronbach’s alpha (CA) coefficient of ≥ 0.6 and a total variable correlation coefficient of > 0.3 (Carpenter 2018).

In exploratory factor analysis (EFA), the criteria for sample size appropriateness include a Kaiser-Meyer-Olkin (KMO) coefficient ranging from 0.5 to 1, a significant Bartlett test (p < 0.05), and a factor loading of at least 0.3 (Taherdoost et al. 2022). A loading factor of 0.5 indicates good statistical significance, with a total extracted variance ≥ 50%. Eigenvalues greater than 1 were retained.

Confirmatory factor analysis (CFA) assesses the fit of a measurement model to actual data, enhancing reliability and validity and refining model suitability (Mishra 2016). CFA is a powerful statistical technique used to assess the fit between a previously constructed theoretical model (including latent factors) and the actual data collected. In a research context, CFA plays a key role in validating the structure of a scale. The main purpose of CFA is to confirm the validity of the scale, ensuring that the indicators used truly represent theoretical factors such as product quality, price, and sales skills, thereby forming broader concepts such as “loyalty.” The results of CFA analysis are evaluated through fit indices such as CFI, TLI, RMSEA, and GFI.

Structural equation modeling (SEM) is a second-generation statistical technique used to analyze multidimensional relationships between variables. It visually represents these relationships and enhances theoretical predictions by specifying measurement properties and the relationships between latent variables (Haenlein and Kaplan 2004; Brown and Moore 2012). SEM allows researchers to simultaneously test complex causal relationships between multiple variables, including latent variables. In research, SEM is applied to test and quantify the impact of factors such as price, product quality, brand reputation, etc. on target variables such as customer loyalty. The significance of the results of SEM analysis is expressed in the statistical significance of the relationships (usually assessed by the p-value, with a common threshold of p < 0.05) and the sign of the regression coefficient.

To conduct EFA, a large sample size is required and is determined based on the minimum sample size and the number of variables included in the analysis. With a minimum observation/variable ratio of 5:1 and 32 questions in the adjusted survey, the calculated minimum sample size was 160 (Carpenter 2018). In this study, the actual sample size was 747, meeting the sample size requirements.

Convenient random sampling was employed using survey methods through direct phone interviews, Google Forms, or printed questionnaires with customers who purchased medicines at pharmacy chains or met the sampling criteria in Can Tho City during the 2023–2024 period. Participants completed surveys based on their real-life experiences purchasing and using services at these chains (Huang et al. 2021).

Data were analyzed using SPSS 26.0, using descriptive statistical data analysis to calculate frequency, mean, and standard deviation. Correlation analysis was also performed to assess the degree of correlation between variables. Finally, AMOS software was used for moderation-mediation analysis, which included regression calculations, CFA, and SEM (Stepurko et al. 2016; Van Fleet and Peterson 2016; Zhou et al. 2017).

PQ Product quality

BN Brand name

P Price

F Facilities

C Convenience

ACP Advertising and consumer promotion

SS Selling skills

L Loyalty

CA Cronbach’s alpha

TLI Tucker-Lewis index

CFI Comparative fit index

GFI Goodness of Fit index

RMSEA Root mean square error approximation

PCLOSE Probability of close fit

Table 1 shows that most study subjects are female (62.2%), while male customers make up a smaller proportion (37.8%). Notably, a high proportion of respondents (80.7%) indicated that pharmacy chains are their first choice for medical needs.

As shown in Table 2, several factors were eliminated due to their impact on the CA coefficient. Factors P2, C2, ACP3, and L4 were removed because the CA coefficient increased after their removal (0.991 > 0.989; 0.999 > 0.983; 0.966 > 0.963; 0.977 > 0.975). In contrast, the remaining factors were retained because they met the required conditions.

Cronbach’s alpha values for CSR, trust, and customer loyalty (including loyalty and common behavior) are presented in Table 3.

The extracted variance value surpassed 70%, reaching 93.839%. Simultaneously, KMO = 0.886 (0.5 ≤ KMO ≤ 1) and Bartlett’s test with sig. = 0.000. The six groups of factors extracted at Eigenvalue = 1.014 (>1) all meet the conditions (Table 4). During the rotated exploratory factor analysis (EFA), a new factor emerged, consisting of items from multiple original subsections (SS1, SS3, SS4, SS2, KE1, KE2, KE3, C1, and C3). This factor was not labeled due to its mixed composition. Meanwhile, the remaining factors reorganized items from their original subsections into corresponding new constructs, and these factors were not renamed. Therefore, the variables retained after EFA analysis fully satisfy the condition that the loading factor is greater than 0.5, indicating their significant role and practical meaning.

Table 5 indicates that the analysis results show satisfaction with a KMO value = 0.910, Bartlett’s test with sig. = 0.000, and an Eigenvalue = 5.385 (>1), meaning that one factor was extracted. The total variance extracted was 89.742% (>50%). As a result, EFA analysis was conducted with the observed variables of the dependent component “(L): Loyalty.

According to Fig. 2, the model appears to be a good fit for the actual data, with the following good fit indices: TLI = 0.975, CFI = 0.981, GFI = 0.916 (>0.9), RMSEA = 0.057 (<0.08), PCLOSE = 0.002, and Chi-square/df = 3.414 (<5). All standardized factor loadings met the criteria, indicating that each observed variable significantly contributed to its corresponding construct. These results confirm that the questionnaire reliably measures the underlying theoretical constructs. The good fit indices suggest that the proposed factor structure is valid and appropriate for further analysis. This supports the unidirectionality of the measurement scales, as explained by Hu and Bentler (1999) in their work on cutoff criteria for fit indexes in covariance structure analysis. This means that the constructs identified in the study (e.g., product quality, brand name) are well-defined and consistently measured, ensuring the reliability of the survey instrument.

Figure 2. 

Description of CFA results.

Fig. 3 shows that the model result appears to be a good fit for the actual data, with the following fit indices: GFI = 0.895, CFI = 0.981, TLI = 0.976 (all indices greater than 0.9), RMSEA = 0.057 (<0.08), PCLOSE = 0.002, and Chi-square/df = 3.401 (Hu and Bentler 1999).

Figure 3. 

SEM testing results.

The results depicted in Table 6 show the impact factors as follows: The analysis indicated that brand name (BN), price (P), facilities (F), convenience (C), knowledge of employees (KE), and selling skills (SS) all have positive and statistically significant effects on customer loyalty (L), with significance levels of p < 0.05.

Our research showed that a higher proportion of females (62.2%) in the survey population compared to males (37.8%). This finding is similar to Adat’s research on customer satisfaction in retail pharmacy chains in Durban (Adat 2013). The similarity can be attributed to the tendency of women to shop for cosmetics, drugs, and other medical supplies for their family. Most survey participants have an intermediate level (47.9%). This suggests that consumers generally possess basic knowledge about pharmaceuticals at pharmacy chains, but their understanding is not in-depth. The reason is that consumer knowledge is not scientifically proven due to differences in perspectives between consumers and drug manufacturers. In light of this, policymakers should raise social awareness by providing comprehensive information to consumers (Mohammadzadeh et al. 2017).

According to the survey results, we have identified several factors that, in line with existing literature (Mircheva and Pelev 2012), influence customers’ return to retail pharmacy chains, which are discussed in detail below.

Regarding product quality, most survey participants believe that the products sold at pharmacies have a shelf life of over six months and demonstrate good treatment effectiveness. However, a small portion of participants feel that the products at the pharmacy do not fully meet their needs. Various activities, including professional drug guidance and consulting services, have been implemented to provide buyers with comprehensive information. These efforts aim to enhance consumers’ understanding of the quality of products sold at pharmacy chains (Liu et al. 2021).

The brand of the pharmacy chain significantly influences customer attraction and retention. Nearly all survey participants trust the quality of products associated with the pharmacy chain’s brand. They find it easy to access information on the Internet, leading to a heightened sense of security when choosing a branded pharmacy. Customers have trust in the pharmacy chain’s brand because they believe that a product sold at a branded pharmacy chain has acceptable quality, which aligns with a study by Mohammadzadeh (Mohammadzadeh et al. 2017). Brand image is an asset that develops through the relationship between the pharmacy chain and its customers, so most customers will make purchasing choices based on product availability and brand image (Reinert and Murray 2017). This is demonstrated by successful examples such as Long Chau, Trung Son, and other well-known retail pharmacy chains in Can Tho City. Therefore, retail pharmacy chains should develop brand strategies to attract and retain a large customer base.

In terms of facilities and convenience, the pharmacy chain with airy space, display shelves with fully arranged price lists, a shopping website, a return policy, and the geographical location of the pharmacy chain are all elements that survey participants find highly appealing. The physical environment and all tangible aspects have a strongly positive influence on satisfaction and experience (Brandão and Ribeiro 2023). Promotional advertising policies are also regarded as a tangible feature since they can offer clients information about the type of service they expect and can play a vital role in enhancing customer satisfaction and generating long-term profits. Other tangible elements include consultation rooms, equipment, and promotional materials in a retail pharmacy chain (Adat 2013).

Price significantly impacts customer return in pharmacy chains, as it directly impacts their subconscious perceptions of product value. Survey respondents agree that lower prices increase return likelihood and compare prices between stores to prefer those with reasonable pricing. Therefore, pharmacy chain businesses should develop a pricing strategy that contributes to building the pharmacy chain’s brand image, encouraging customers to make repeated purchases (Reinert and Murray 2017).

Most participants express satisfaction with pharmacy chain employees’ knowledge and sales skills, particularly in providing advice on drug prices, quality, dosage, usage, and potential side effects, and show interest in their attitude and consulting skills. Hung found that, in situations where there was no opposition, people had a high belief in pharmacists’ competence to prescribe antibiotics without a prescription and that they had the essential skills for patient consultation and administering antibiotics when needed (Hung et al. 2023). For that reason, this indicates that customers’ trust in pharmacists has a highly positive influence on trust in the pharmacy chain.

Castaldo asserts that trust in community pharmacists is the primary driving force behind customer satisfaction. This trust not only directly influences satisfaction levels but also plays a crucial role, either directly or indirectly, in building trust in pharmacy chains (Castaldo et al. 2016). Therefore, pharmacy chains should enhance their staff’s competencies, skills, and attitudes by carefully selecting and training employees. Mohammadzadeh also found that customer satisfaction increases when community pharmacists spend more time on consultation services (Mohammadzadeh et al. 2017).

When considering customer loyalty, the process of building loyalty is intricately tied to the quality of pharmacy services. The presence of a knowledgeable employee with a positive attitude plays a crucial role in enhancing customer satisfaction, making them more inclined to return for future purchases (Mircheva and Pelev 2012). Additionally, fulfilling customer needs by delivering excellent service quality is recognized as a key factor in fostering customer loyalty. This not only fosters customer satisfaction but also significantly contributes to shaping the image, reputation, and brand of pharmacy chains (Castaldo et al. 2016). Customer satisfaction is part of the brand experience and can lead to loyalty and long-term relationships between the brand and its customers. These factors interact with each other, contributing to customer satisfaction with the pharmacy chain (Reinert and Murray 2017).

These factors positively influence customer returns to pharmacy chains in Can Tho City. It is imperative for pharmacy chains to develop comprehensive plans and policies. Neutralizing these factors will ensure the future success and development of pharmacy chains.