This study was a cross-sectional study. The survey was conducted in June 2023. The study population was customers of community pharmacies in Magelang. The sampling technique used in this study was purposive sampling method. The samples were used in this study of 252 respondents. The inclusion criteria were: participants know the terms and conditions and are willing to be research respondents, more than 17 years old, pharmacy customers who had visited at least two times and willing to participate in the study.

The results of the outer model analysis in Table 2 and Table 3 show that the measurement model has met the convergent and discriminant validity requirements.

The loading factor is a parameter used to indicate the suitability level of an item explaining a variable. An item can explain a variable very well if the loading factor value is more than 0.70. However, in range 0.50 to 0.60 is tolerable. AVE is used to measure the number of variances that can be compared to variances caused by measurement errors. The AVE value must be greater (>0.5) because this value represents a convergent validity which means that one latent variable is able to explain more than half the variants of its indicators in the average. Composite Reliability can be used to measure the true reliability value of a model. The composite reliability value should be greater than 0.7 but a value of 0.6 is acceptable.

Discriminant Validity can be done by comparing the square root of average variance extracted (AVE) value with the correlation between other variables in the model, so the discriminant validity value is said to be good if the AVE value is above 0.5.

This model focuses on the latent variable structure model. The inner model is the part of the model that describes the relationships among the latent variables that make up the model. There are 3 core models in this study, namely Model I (satisfaction), Model II (trust in pharmacist), and Model III (trust in pharmacy). The model and the results of hypotheses testing are shown in Table 4 and presented with different outputs in Fig. 1, but both have the same interpretation. Model I shows the magnitude of R2 (coefficient of determination) is 0.443, these value indicates that 44.3% of the variance of satisfaction can be explained by variance of the five independent variables namely infrastructure, communication, medication information, product availability, and trust in pharmacists while the remaining 55.7% is explained by other variables outside the model. Model II shows that the magnitude of R2 is 0.303, this value indicates that 30.3% of the variance in trust in pharmacists can be explained by infrastructure, communication, medication information, and product availability while the remaining 69.7% can be explained by other variables outside the model. Model III shows the magnitude of R2 is 0.546. This value indicates that 54.6% of the variance in trust in pharmacy can be explained by satisfaction and trust in pharmacists while the remaining 45.4% is explained by other variables outside the model.

Figure 1. 

Path coefficients and t-statistics.

The infrastructure had a significant effect on satisfaction and trust in pharmacist (p<0.05), these findings are in accordance with research conducted by Castaldo (Castaldo et al. 2016). Other factors that can also affect customer satisfaction include facilities, clean location, and the comfort of the waiting room (Pribadi et al. 2019). The convenience and atmosphere of the store have an important role in predicting customer experience. Pharmaceutical retailers must also consider target consumers in determining store attributes (Jack and Ling 2016). Communication had non-significant effect on satisfaction and trust in pharmacist (p>0.05). This finding is certainly very surprising because it is not consistent with the results of Pribadi, Castaldo, and Wang studies (Castaldo et al. 2016; Wang et al. 2018; Pribadi et al. 2019). Many studies show that nonverbal communication is as important as verbal communication. The poor relationship between the pharmacist and the patient is the result of poor nonverbal communication Provider-patient communication is a factor that contributes to patient satisfaction, adherence, and health outcomes (Wang et al. 2018). Research conducted by Antari et al. which examines the factors that influence the level of trust when observed in terms of communication, they found the empathy and closeness affect the level of patient trust (Antari et al. 2019). Research by Perrault indicates that many patients appreciate the active role of pharmacists in their care, most patients prefer to take the less active role (Perrault and Beal 2018). According to Gordon et al, improving communication is more valuable and able to affecting health outcomes of patients (Gordon et al. 2015).

Medication information had a significant effect on satisfaction and trust in pharmacist (p<0.05).This finding consistent with previous studies Khudair, Larson, and Panvelkar (Larson et al. 2002; Panvelkar et al. 2009; Fahmi Khudair and Raza 2013). According to Tjong, consumer trust towards types of information from pharmacists included advice on non-prescription drugs, how to take drugs, and how to deal with minor illnesses such as headaches, coughs, and colds, diarrhea, constipation, and hemorrhoids, However, information regarding herbal products are still unknown by consumers (Allam et al. 2014). Another study conducted by Lexcin et al. found that the types of information trusted by consumers include information on how to consume drugs, side effects of drugs, and the price of drugs purchased by prescription from a doctor (Lexchin and Mintzes 2002) Research conducted by Abdullah et al. Stated that the majority of pharmacy customers believe that drug information services can be beneficial for consumers. However, pharmacy consumers do not believe in the pharmacist’s ability to provide drug information services. Therefore, to gain the trust of pharmacy consumers, pharmacists must play an active role in providing drug information services in pharmacies (Sridevi et al. 2017).

Product availability had a significant effect on trust in pharmacist (p<0.05). However, it had no significant effect on satisfaction (p>0.05). This finding is contrary to studies by MacKeigan and Larson which found a positive relationship between patient satisfaction and drug supply (Mackeigan and Larson 1989). The availability of over-the-counter (OTC) drugs has a significant relationship with patient satisfaction (Kamei et al. 2001). The context of pharmaceutical services is now expanding not only to the supply of medicines but also patients are now looking for other factors when they visit the pharmacy (Fahmi Khudair and Raza 2013). Therefore, service activities that initially focused on drug management as a commodity must be changed to a comprehensive service that includes providing information to support the correct and rational use of drugs, monitoring drug use, preventing the possibility of medication errors, and patient-centered care to improve the quality of life of patients.

Several limitations in this study include: the samplesize of this study was less representative of the population, the research period is relatively short, R-Square values which were relatively small 44.3%; 30.3%; and 54.6% showed that there are still many other variables outside this study that affect customer satisfaction and trust in pharmacy. Further research needs to consider the use of a larger sample, the use of the longitudinal survey method is likely to provide better results, other independent variables not measured in this study need to be investigated. Emotional factors seem to have an influence on satisfaction and trust.