Article Text
PDF
Abstract
Objective To assess healthcare providers’ intentions and the associated factors to use mobile phone-based short message service (SMS) to support adherence and care of tuberculosis (TB) patients in the Oromia region of southwest Ethiopia.
Study design An institutional-based cross-sectional study was conducted from October to November 2022.
Study setting The study was conducted in public hospitals which include Mettu Karl referral hospital, Dembi Hospital, Bedelle Hospital, Darimu Hospital and Chora Hospital in Ilu Aba Bor and Buno Bedelle zones.
Participants A total of 625 (54.9% male and 45.1% female) health professionals participated in the study. The study participants were selected using a simple random sampling technique. All health professionals permanently working in Ilu Aba Bor and Buno Bedelle zone hospitals were included in this study. However, health professionals with less than 6 months of experience and those who were not present during the data collection period were excluded from this study.
Outcome measure The intention to use mobile phone-based SMS to support TB patients.
Results Healthcare professionals’ intention to use mobile SMS was 54.4%. Effort expectancy had a significant direct effect on attitude (β=0.162, p<0.01) and intention towards using mobile phone SMS (β=0.329, p<0.001). The intention to use mobile phone SMS was directly influenced by facilitating conditions (β=0.104, p<0.01) and attitude (β=0.26, p<0.001). The relationship between effort expectancy and intention to use SMS was mediated by attitude (β=0.043, p<0.01).
Conclusions Overall, intention to use of mobile-based SMS was high. Effort expectancy, attitude and facilitating conditions were significant factors that determined healthcare professionals’ behavioural intention to use mobile phone SMS. Effort expectancy had a more significant prediction power than others. As a result, system forms that are easily interactive and applicable should be implemented to improve capacity building and support the adherence and care of TB patients.
- Tuberculosis
- Information technology
- Telemedicine
- Information management
- Health informatics
Data availability statement
Data are available upon reasonable request. Upon reasonable request from the corresponding author, the data sets created and/or analysed during the current work will be made available.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
STRENGTHS AND LIMITATIONS OF THIS STUDY
This study used the unified theory of acceptance and use of technology, which is the most accurate and recent model to identify important factors to influence mobile health technologies.
The study covers public hospitals with a large sample size, which improved its generalisability.
This study may be tilted in favour of social desirability because it is a cross-sectional survey.
The study was not supported by qualitative findings.
The study did not include private hospitals.
Introduction
Globally, it is crucial to cure tuberculosis (TB) to reduce morbidity, mortality and the risk of continued transmission, but it can be difficult to support and monitor patients during the entirety of treatment.1 2 The consequences for patients who experience treatment interruptions with TB therapy include drug resistance, recurrence and death.2–4 The WHO recommends in-person, directly observed therapy, either at home or in a medical facility, to increase TB adherence.5 However, directly observed therapy is frequently difficult and impractical to implement in most areas with a high TB burden.6
According to studies carried out in northwest Ethiopia, 34% of patients’ arguments for not adhering to TB treatment were forgetfulness.7 8 Community members may be too preoccupied with their daily tasks to remember and monitor patients for such a long time. Thus, the introduction of digital health technologies by healthcare professionals may improve patient adherence to TB therapy.
Digital technology adoption is viewed by health professionals as a key method for improving the efficacy, efficiency and quality of healthcare delivery because it can improve patient care quality and service delivery efficiency while minimising medical errors.9–11 For instance, mobile technology’s short message service (SMS) feature can be used to track upcoming appointments, preventative immunisations and long-term self-management treatments.9 To encourage the development of digital health innovations in worldwide efforts to improve TB care and prevention, the WHO has already formed its global task force on digital health, such as mobile health.12 13
With extensive adoption in low-income and middle-income countries, especially those with lower socioeconomic standing, the number of mobile phone subscriptions now outnumbers the number of people worldwide.14 Individuals now have own smartphones, making it possible in some locations with a strong internet connection to conduct video conversations. However, in comparison to other developed regions, the use of mobile technology in the healthcare system in developing countries is too often limited to certain roles for health interventions. The evidence done in Ethiopia showed that mobile phone messaging revealed a positive effect in improving healthcare services,2 13 15 but the intention level of healthcare providers to use mobile health with SMS to support patients in the resource-limited setting is uncertain.
Theoretical background and hypothesis development
Different types of models have been used to identify factors associated with the acceptance and use of health information system technologies.16 17After looking at eight models, including the theory of reasoned action, the technology acceptance model, the theory of planned behaviour, the social cognitive theory, the motivational model, the combined technology acceptance model and the innovation diffusion theory; the Unified Theory of Acceptance and Use of Technology (UTAUT) model was developed to put forth a unified theory of technology acceptance and it is the most accurate and up-to-date technology acceptance model used to evaluate intention to use and actual use of technology.16 18
The acceptance and utilisation of mobile-based SMS for TB treatment monitoring depend on the particular patients, healthcare providers and facilities that serve TB patients. The UTAUT model is suitable for this study due to its robustness and broad use in analysing technology adoption at the individual level. Accordingly, we used the model by adapting or modifying the constructs in our study setting
Accordingly, after we used the fundamental constructs of the UTAUT model (performance expectancy (PE), effort expectancy (EE), social influence (SI) and facilitating condition (FC)), we modified it in the context of a comprehensive review of the literature.16 19–21 So, attitude towards using technology (ATT) may have a substantial impact on the intention to use information technology (behavioural intention, BI) in developing nations. Actual use is another factor in the UTAUT model that is important when the solution is frequently accessible. However, this variable was not included in our study because the solution is still not accessible for continuous use (figure 1). The constructs used and the associations that were hypothesised and tested in this study are provided below.
Modified UTAUT models of healthcare professionals’ intention to use mobile phone short message service and its predictors for adherence support and care of patients with tuberculosis infection in a resource-limited setting, 2022. UTAUT, Unified Theory of Acceptance and Use of Technology.
Behavioural intention to use (BI): it is the degree to which a person has made deliberate plans to engage in or refrain from engaging in a particular future action.18 In this study, healthcare professionals are questioned about their intentions to employ the proposed solution when it becomes accessible in the future.
Performance expectancy (PE): respondents were questioned about the solution’s value and how it influenced their task effectiveness, productivity, job effectiveness, quality of work and the TB monitoring process. Thus, healthcare professionals are more likely to use mobile phone-based SMS, if they think that PE is high.9 22–26 Based on the above-mentioned literature, the following hypotheses were investigated in this study:
H1PE has a positive influence on the intention to use mobile phone SMS technology.
H2: PE has a positive influence on attitudes towards using mobile phone SMS technology.
Effort expectancy (EE): system design has to do with how simple it is to use, learn and build new skills, and comprehend how the suggested solution interacts with the user. In the treatment value chain, healthcare professionals are expected to enter information, read it and engage with communication technology like mobile phones. According to studies, users’ intention to utilise technology is influenced by their effort expectations.9 23 24 26 Based on these, the following hypotheses were investigated in this study:
H3: EE has a positive influence on the intention to use mobile phone SMS technology.
H4: EE has a positive influence on attitude to use mobile phone SMS technology.
Social influence (SI): this plays a significant role in the adoption and continued use of the solution given the experience of inconsistent adherence to reporting guidelines and treatment regimens by patients, medical professionals, allied healthcare professionals and organisations. Intention to use the mobile phone for SMS may therefore be strong, if SI is positive.9 22 23 26 In the context of this, the following hypotheses were investigated in this study:
H5: SI has a positive influence on the intention to use mobile phone SMS technology.
H6: SI has a positive influence on attitudes towards using mobile phone SMS technology.
Facilitating condition (FC): conditions in the context of TB treatment monitoring include having the knowledge and tools required to utilise the system as well as the proposed solution’s compatibility with existing systems.9 22 25 26 On account of these empirical findings, our study makes the following hypothesis:
H7: FC has a positive influence on the intention to use mobile phone SMS technology.
H8: FC has a positive influence on attitudes towards using mobile phone SMS technology.
Attitude (ATT): in a context with low resources, health professionals have little knowledge about access to technology, which makes changing their attitude about using a new technology.16 We reasoned that including attitude as a category could seem pertinent to studying behavioural intentions to use new technology in certain contexts.16 22 24 27–30 On account of those findings, our study makes the following hypothesis:
H9: Attitude has a positive influence on the intention to use mobile phone SMS technology.
H10: Attitude mediates the relationship between PE and intention to use mobile phone SMS technology.
H11: Attitude mediates the relationship between EE and intention to use mobile phone SMS technology.
H12: Attitude mediates the relationship between SI and intention to use mobile phone SMS technology.
H13: attitude mediates the relationship between FC and intention to use mobile phone SMS technology.
The main beneficiaries of this study will be patients, health professionals, Regional Health Bureaus and health organisations for health system practices. According to our search of the literature, little research has been done on the subject of healthcare professionals’ intentions to use mobile phone SMS using the UTAUT model in Ethiopia. As a result, this study investigates, introduces and empirically tests a modified theoretical model based on the UTAUT model to identify the main factors influencing healthcare professionals’ intention to adopt a mobile phone SMS system.
Methods and materials
Study design and setting
The institutional-based cross-sectional study design was conducted from October to November 2022 at public hospitals in the Ilu Aba Bor and Buno Bedelle zones, Oromia Regional State, southwest Ethiopia. Illu Aba Bor Zone comprises one town administration and 14 rural districts with a total population of 1 606 502, according to the 2019 District Finance and Economic Development Office annual data report. Among those zones, five hospitals exist, namely Mettu Karl Referral Hospital, Dembi Hospital, Bedelle Hospital, Darimu Hospital, and Chora Hospital.
Study population
All healthcare professionals working in the southwest Oromia region of public hospitals in the two zones were used as study populations. All health professionals, working at public health hospitals, who were available during data collection time were included in the study. However, health professionals with less than 6 months of working experience and who were on maternal or annual leave, or those who were seriously ill during the study period were excluded from the study.
Operational definition
Intention to use mobile-based SMS: the extent to which a person has made intentional plans to use new technology.31 32 When a healthcare professional rates intention to use a technology measurement and scores median, and above the median was considered as intended to use mobile phone-based SMS, else not intended to use it, with a five-point Likert scale of three questions.31
Sample size determination and sampling procedure
The sample size was estimated based on assumptions of determining model-free parameters via the modified model31 32 (figure 1). By considering 25 variances of the independent variables, 6 covariance between independent variables, 16 load factors between latent and latent indicators and eight direct effects and four indirect effects of regression coefficients were estimated. Finally, a total of 59 free parameters were estimated. But the variances of dependent variables, the covariance between dependent variables and the covariance between dependent and independent variables are never parameters.33 34 Accordingly, a 1: 10 ratio of responders to free parameters to be estimated was suggested to estimate the sample size based on the number of free parameters in the hypothetical model.31 32 35 As a result, the minimum sample size was 590, through taking participants with a free parameter ratio of 10, since the computed sample size considers the 10% non-response rate. Thus, the final sample size of 649 study participants was calculated.
The number of healthcare professionals from the five hospitals was calculated using population proportional allocation and a simple random sampling technique. Finally, consecutive healthcare professionals who fulfilled the inclusion criterion were included in the study until the allocated size was obtained in each of the five hospitals.
Data collection tools, data quality control and procedures
To investigate the hypotheses provided by this study, a structured questionnaire was used from earlier investigations.16 27–30 There are two sections to the questionnaire. Section A: focuses on user demographic data; Section B: contains 18 positive statements that symbolise the UTAUT constructs. A five-point Likert scale, with one denoting strongly disagree and five denoting strongly agree, was used to measure the constructs.
To control the quality of the data, 2 days of training were given to data collectors and supervisors on the objective of the study, data collection procedures, data collection tools, respondents’ approaches, data confidentiality and respondents right before the data collection date. Before the actual data collection, pretesting of the questionnaire was conducted among 10% of the study participants outside of the study (Jimma Hospital). After obtaining feedback from the respondents, language experts modified the wording of the questions.
The internal consistency of the items was estimated using the Cronbach alpha coefficient (Cα), and composite reliability (CR). The results showed that all of the items’ scores were above 0.7. The assumptions were examined for outliers, multicollinearity and independent error factors before running the structural equation model. The results showed that there was no existing multicollinearity, with the overall variance inflation factor (VIF) value being less than 2 and the tolerance being greater than 0.5.
Patient and public involvement
Patients and public were not involved
Data processing and analysis
Data from respondents were entered into Epi-Data V.4.6 and exported to SPSS V.25 for descriptive data analysis. Additionally, model constructs were evaluated using structural equation modelling (SEM) analysis via Analysis of Moment Structure V.26 software. To test measurement model, confirmatory factor analysis (CFA) using standardised data was used. As part of CFA, a correlation between constructs less than 0.8 and factor loadings greater than 0.6 for each item was examined.36 The χ2 ratio (<5), the Tucker-Lewis’s Index (TLI>0.9), the Comparative Fit Index (CFI>0.9), the Goodness of Fit Index (GFI>0.9), the Adjusted Goodness of Fit Index (AGFI>0.8), the root means square error approximation (RMSEA<0.08) and the root mean square of the standardised residual (RMSR<0.08) were all used to evaluate the goodness of fit.37–39
Data normality was evaluated using multivariate kurtosis<5 and the critical ratio (CR) between −1.96 and +1.96. Multicollinearity was also tested using VIF<10 and tolerance>0.1, as well as the correlation between exogenous constructs of <0.8, and there were no problems. In the measurement model, Cronbach’s alpha test was used to assess construct reliability; each construct in the study met the required threshold of >0.70, and the CR was above 0.70.40 The Average Variance Extracted (AVE) approach, with values above 0.5, was used to assess the converging validity, while the square root of AVE in Fornell Larcker criterion, with values below 0.9, and the correlations with other latent constructs should have a lower value than the square root of AVE in each construct was used to assess the diverging validity.31 36 38 41
To evaluate a structural model, the CR and the path coefficient were used to analyse the relationship between exogenous and endogenous variables. A p-value of less than 0.05 and 95% CIs were used to assess the predictors’ statistical significance. The influence and level of significance of each of the four possible mediation paths in the model were explored, partial mediation occurs when a construct’s direct, indirect and total effects are all statistically significant, and full mediation occurs when the direct and indirect effects are both statistically significant but the total effect is not. To confirm mediation, we typically looked for a substantial indirect effect with a p-value<0.05.
Results
Sociodemographic characteristics of healthcare professionals
A total of 625 study participants were included in the study with a response rate of 96.3% of them who gave their consent and responded to the questions. Of the total of (n=625) respondents, 343 (54.9%) of them were males, 293 (46.9%) respondents were in the age group of <30 years and the mean age of the respondents was 32±6.01 years. About more than half of the respondents 451 (72.2%) had less than or equal to 3 years of work experience. In addition, 228 (36.5%) of the respondents were medical doctor professionals (table 1).
Sociodemographic characteristics of healthcare professionals in a resource-limited setting, 2022 (n=625)
Intention to use mobile-based SMS
Overall, 340 (54.4 %; 95% CI: 49.9 to 58.1]) healthcare professionals were intended to use mobile phone-based SMS in southwest Ethiopia. The median score of intention to use mobile phone-based SMS was 10.1, with a SD of 3.3. The minimum and maximum scores were 3 and 15, respectively.
Measurement model
Reliability and validity of the construct
All of the constructs had Cronbach’s alpha scores between 0.77 and 0.91 and CR scores between 0.78 and 0.92, factor loadings of the items ranged between 0.64 and 0.89, and the value of AVE varied between 0.61 and 0.71. Therefore, all of the constructs had strong convergent validity (online supplemental table 1).
Supplemental material
According to the findings, the bolded values (diagonal values) or the square root of the AVE of the construct were higher than the other values in that column, and the raw. As a result, the model’s constructs’ discriminant validity has been proven (online supplemental table 2).
Supplemental material
Goodness of fit
The results in CFA showed that model fit indices with respective values were χ2 difference (x2/df=2.76), GFI=0.93, AGFI=0.91, CFI=0.94, TLI=0.93, root means the square error of approximation (RMSEA=0.05) and standardised root mean squared residual (SRMR=0.03). Accordingly, the goodness of fit model’s values met the requirements.
Structural equation model
According to the result, EE had a significant direct effect on healthcare professionals’ attitude to using mobile phone SMS (β=0.162, 95% CI: 0.061 to 0.270, p<0.01) and intention towards using mobile phone SMS (β=0.329, 95% CI: 0.233 to 0.433, p<0.001). The FC had a direct significant effect on healthcare provider’s intention towards using mobile phone SMS (β=0.104, 95% CI: 0.025 to 0.191, p<0.01), and also healthcare professionals’ attitude towards the system had a direct significant effect on behavioural intention to use mobile phone SMS (β=0.268, 95% CI: 0.163 to 0.373, p<0.001), whereas the relationship between PE (β=0.268, 95% CI: −0.017 to 0.170, p=0.109), SI (β=0.015, 95% CI: −0.044 to 0.075, p=0.610) and FCs (β=0.055, 95% CI: −0.026 to 0.140, p=0.184) were not significant effect on healthcare professionals’ attitude towards using mobile phone SMS, and the path relationship of PE (β=0.002, 95% CI: −0.088 to 0.094, p=0.980) and SI (β=−0.030, 95% CI: −0.086 to 0.028, p=0.318) to use mobile phone SMS were not statistically significant (table 2).
SEM analysis of healthcare professionals’ intention to use mobile phone short message service and its predictors for adherence support and care of patients with tuberculosis infection in a resource-limited setting
According to the finding, PE, EE, SI and FCs accounted for 68% and 74% of the variance (R2) in attitude and intention to use mobile phone SMSs, respectively (figure 2).
SEM analysis of healthcare professionals’ intention to use mobile phone short message service and its predictors for adherence support and care of patients with tuberculosis infection in a resource-limited setting, 2022. SEM, structural equation modelling.
Mediation effects
The results revealed that attitude played a mediating role in the relationship between healthcare professionals’ effort expectations and their intention to use mobile phone SMS to support TB patients, as opposed to the relationship between PE, SI and FCs. Both the relationship between attitude and intention to use the system, as well as the regression coefficient between attitude and EE, were statistically significant. The indirect effect of the standardised estimation value was 0.043 and p<0.01. In the context of this, the relationship between EE and an intention to use mobile-based SMS was statistically significant through attitude (table 3).
Mediating effects of attitude and healthcare professionals’ intention to use mobile phone short message service and its predictors for adherence support and care of patients with tuberculosis infection in a resource-limited setting
Discussion
This study was to determine the level of intention and its predictors that may influence the use of mobile phone SMS systems to support adherence and care of TB patients. Overall, 340 (54.4%; 95% CI: 49.9 to 58.1) of healthcare professionals intended to use mobile phone-based SMS to support TB patients in southwest Ethiopia. This finding was higher than the study done on the intention of health providers’ electronic medical records in the Amhara regional state of Ethiopia (39.8%).42 This discrepancy could be due to the study period and the use small sample size. Similarly, the willingness of healthcare professionals to use mobile-based health services in Ethiopia was 44%.43 This possible difference could be due to the study participants only obstetric health providers and they used a smaller sample size.
The findings show that EE, attitude and FCs were significant factors that determined healthcare professionals’ behavioural intention to use mobile phone SMS. EE also influences healthcare providers’ attitudes towards using these systems. Furthermore, the attitude acted as a partial mediator between effort expectation and intention to use the system. Thus, H3, H4, H7, H9 and H11 were supported and thus can be used as effective measures of mobile phone SMS adoption in resource-limited settings.
According to the findings, healthcare professional EE had a positive direct effect on attitude and both direct and indirect effects on the intention to use mobile phone SMS. This suggested that when healthcare professionals perceived the system’s simplicity or lack of effort in use, their perceptions of its usefulness, attitude and intention to use mobile phone SMS were significantly improved. This finding is consistent with findings from other studies in different countries.9 24 44–46 This could be because an individual’s attitude and acceptance of the use of mobile phone SMS systems are greatly affected by the work required to manipulate the system and if the system is expected to manipulate people’s intentions to use mobile phone SMS systems with less effort, the system’s performance will improve.47 As a result, when implementing the use of mobile phone SMS technologies, the system should be simple for healthcare providers to understand and use to ensure long-term system adoption.
FCs also positively influenced the intention to use mobile phone SMSs to support adherence and care of TB patients. This showed the significance of creating the organisational and IT requirements for this example, including the resources, technical support and training that are necessary for the adoption and ongoing use of mobile-based SMS-enabled solutions by healthcare providers in the public health system This finding is consistent with findings from other studies in different countries.9 25 26 48 This could be because health professionals think that as mobile phone use has increased, more people have access to them and are familiar with using devices to communicate with their intended recipients. As a result, they are convinced and have no trouble supporting patient care and adherence.
Health professionals’ attitudes positively influenced their intention to use mobile phone SMS. Medical experts supported the usage of mobile phone SMS as a tool for bettering their health and the standard of TB treatment as they observed it being done. This finding is in line with findings from other studies done in different settings.9 22 49–51 Therefore, it is important to give priority to initiatives that change attitudes, such as providing mobile devices at work, providing ongoing training and assistance and fostering knowledge exchange about eHealth technologies such as mobile SMS for the remainder of the patient’s care.
Accordingly, this study provides theoretical and practical implications; the findings might alleviate any concerns regarding mobile-based SMS and its acceptability in resource-constrained environments. MHealth and other digital communication technologies have a positive impact on healthcare. The study also offers a framework for future interventional research aimed at developing and evaluating mobile text messaging interventions as a means of enhancing TB prevention programme in Ethiopia. Policymakers, healthcare providers and planners should be concerned about improving the ability of users by providing technological resources, technical support and training to adopt mobile-based SMS in Ethiopia.
Conclusions
Overall, healthcare professionals’ intention to use mobile-based SMS was high. Effort expectancy, attitude and Facilitating conditions were significant factors that determined healthcare professionals’ intention to use mobile phone SMS. Effort Expectancy also influences healthcare providers’ attitudes towards using these systems. Furthermore, the attitude acted as a partial mediator between effort expectations and the intention to use the system. Among the five influencing predictors, Effort expectancy had a more significant prediction power for healthcare providers’ intention to use mobile phone SMS systems. Therefore, easily interactive and applicable forms of the system should be implemented for healthcare professionals, as well as improving capacity and building on the simplicity of technology to support adherence and the care of TB patients. Finally, we recommend that upcoming investigators include external variables, and all health institutions, similarly, support it with a qualitative study.
Data availability statement
Data are available upon reasonable request. Upon reasonable request from the corresponding author, the data sets created and/or analysed during the current work will be made available.
Ethics statements
Patient consent for publication
Ethics approval
This study involves human participants. Ethical clearance was obtained from Mettu University’s ethical review committees with approval number MeU/RAC/1036/15. Participants gave informed consent to participate in the study before taking part.
Acknowledgments
The authors would like to thank Mettu University's ethical committee for the approval of ethical clearance, and next goes to teaching hospital management, data collectors, supervisors and study participants.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Contributors ADW contributed to conceptualisation, data curation, formal analysis, funding acquisition, investigation, methodology, project administration, resources, software, supervision, validation, visualisation and writing-original draft. AWD did the conception and design, acquisition of data or analysis and interpretation of data, and MKH contributed the funding acquisition, investigation, methodology and resources. ADW and AWD wrote the final draft of the manuscript, and the final draft of the manuscript was read, edited and approved by all authors.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting or dissemination plans of this research.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.