Assessing the acceptability of technological health innovations in sub-Saharan Africa: a scoping review and a best fit framework synthesis

Of all the articles included (n = 38) that aimed to assess the acceptability of a health innovation, only seven defined acceptability (all in different ways) (Table 1).

Only five papers used a theory or framework to measure and understand acceptability (Table 2). These were used a priori (to guide data collection) in four articles and a posteriori (to guide data analysis) in one. The notion of acceptability was often assimilated or swapped with other concepts such as perception, satisfaction, adoption, use, willingness to use, etc. Many articles present results relative to the acceptability of a health innovation without even defining precisely what they mean by acceptability and therefore what was being studied. The results of our scoping review confirm the lack of clear theoretical or conceptual definitions and foundations for measuring and empirically understanding the acceptability.

To measure and understand acceptability, 15 studies used qualitative methods, six used quantitative methods, eight used mixed methods, and for nine articles the method was not specified. The authors documented most often the opinion of health workers (n = 29; some studies had several target populations), followed by those of patients for 21 of the studies. Finally, the acceptability of managers (n = 7), stakeholders in general (n = 4), community members who were not necessarily patients at the time of the study (n = 3), other medical staff (pharmacists, laboratory technicians, etc.) (n = 3), and researchers (n = 1) was also considered. In two studies, the innovation was considered acceptable, but it was not clear for whom this acceptability was assessed. In the majority of articles that presented results, acceptability was described as generally positive.

Theoretical articles on acceptability in the literature distinguish between three main stages in acceptability: before the introduction (or before use) of an innovation, during its introduction (or initial use), and after its introduction (or sustained use) [17, 21]. Based on the included studies, we were able to distinguish, for each of these three phases, two potential sub-phases in the acceptability process (Fig. 3). Depending on the period in which acceptability is studied, the influencing factors will not have the same impact on acceptability. Nevertheless, not all of these times are necessarily experienced for each of the innovations introduced (for example, sometimes the innovation is introduced directly into the care routine, or there is no scaling up).

First, at the “conceptual” time, acceptability is measured after explaining the idea or concept of the innovation, but before its introduction or physical presentation. For example, De Haan and al. [46] measured the acceptability of a tool only after giving the participants an information sheet that explained the device. The second stage, the “tangible” (or concrete) time, is a measure of acceptability after having physically presented the innovation but before it was used and integrated into practices. For example, in a study about a new tool for injecting vaccines, the authors stated: “study staff injected vaccine into oranges to demonstrate CPAD [Compact, prefilled, autodisable device] use, as CPAD is not yet licensed for delivery of pentavalent vaccine in the study countries. […]” [44]. The third stage is the “experimental use”. It is when acceptability is measured while the innovation is being used, but not in the way it should normally be used. For example, Ginsburg et al. [55] explain that “we wanted to ensure the application was not used in the clinical care of the patient and that it did not influence the care of the patient in any way. Therefore, we set the oxygen saturation to read 99 percent at all times”. Then, when the innovation is integrated into the care routines, as it should be used normally, it is the time of "routine use". This is the case of a study conducted in Ghana where the acceptability of malaria rapid diagnostic tests (RDTs) was measured after their introduction into care routines [35]. We have identified a fifth time, the time of "sustained use". There, acceptability is measured when the innovation is integrated into care routines during an intervention, but the support and implementation phases are over. For example, Jensen et al. [38] show differences in responses to acceptability questionnaires that were administered before and at the end of the implementation of the intervention. Finally, the last moment of measurement is the time of “scaling up”, when the innovation is diffused beyond its initial place of introduction. The article by Ansbro et al. [32], for example, shows a relative drop in acceptability after the scaling up of a syphilis point-of-care test: “93.8% of pilot HCWs [health care workers] (15/16) and 62.5% of rollout HCWs (15/24) thought patients were somewhat or very accepting of the RST [rapid syphilis test]”. In the context of a scale-up, we can see both the effect of time (evolution between the different phases of the intervention until scale-up) and of the characteristics of the intervention (who implements the intervention and how) on the acceptability of the technology.

We defined perceived complexity as the degree to which an innovation and the behavior(s) required to use it are perceived as difficult to understand and to use/perform. In several articles, we found that health workers thought that: "the technology was very easy or easy to use, all mentioning the limited training required to be able to use the device" [49]. The perceived ease of use has a positive impact on the acceptability of the technology. On the contrary, in other studies, some health workers perceived difficulties in using the innovation: "they did not always act on the lights because they were confused. They felt that the yellow light was reported too frequently when patients appeared well” [45]. Here, experience plays an important role in relation to perceived complexity, highlighting the importance of documenting the timing of the acceptability study, as noted earlier: "HCPs [health care providers] struggled to identify how to change their answer selections but were later able to self-correct with practice" [55].

It is also important to compare the perceived understanding with the actual understanding of the technology. For example, the malaria RDT was considered by the patients as "a generic test able to identify any cause of illness, not just malaria, and expectations that a test should result in a diagnosis, even following a negative result" [35]. Another article points out that some patients wanted to receive antimalarials despite a negative RDT result [33]. Here, healthcare professionals may be pressured by patients to obtain a diagnosis or treatment that technology alone cannot provide. Thus, a poor understanding of the technology, even if not necessarily perceived (respondents thought they understood the innovation), can negatively influence the acceptability of the innovation.

Perceived disadvantages are the amount of effort, risk and cost perceived to be associated with an innovation and its use. These disadvantages can be very diverse, such as increased clinical costs per febrile illness episode [46], painfulness [48], concerns about the safety and hygiene of the device [50], increased administrative constraints [53], increased workload [55], etc. All these disadvantages are distinct from the complications linked to the complexity of using the innovation (perceived complexity). They are the potentially harmful effects caused by the use of the innovation and its integration into routines, and influence the acceptability of the technology negatively. On the contrary, the perception that the use of the innovation will cause very few difficulties and complications for users and beneficiaries can influence acceptability positively.

We define this dimension as the emotions an individual feels about an innovation and its use. Some articles have reported the "enthusiasm" of health workers using the new tool [33] or the "enjoyment" of patients participating in a consultation using the technology [47]. Regarding the introduction of a new diagnostic test, while 90% of participants reported that "they valued or appreciated the rapid test", among those who refused the technology, "the more common reasons for refusing the skin snip biopsy were that they 'did not like the idea'." [48]. This clearly shows the influence of personal feelings on acceptability. The technology can provide a "sense of relief" [49], can be judged "likeable" [50], or a "satisfying experience" [53]. Patients may be enthusiastic about the use of new technologies in health care: "when you use it on the child, the mother becomes excited. She sees that we are doing something new on the child and for her it means the facility is improving" [54]. On the contrary, some may feel fear, related to the pain that the tool could cause [33], or to the fear that the tool will be used for purposes other than those announced [69]. It may also upset the patients' certainties: "the RDT results challenged the patient's medical knowledge and brought into question their (presumed) control over their family's health" [62]. Depending on how people feel about the technology, this can have a positive or negative impact on acceptability.

We defined social influence as the extent to which other people's opinions influence one's degree of acceptability. This dimension took several forms in the empirical data of the articles. From the patients' point of view, the influence exerted by health workers, by their status, was noted: "many respondents conveyed their role as a passive recipient of care, entrusting the clinician with the responsibility to know and prescribe what was 'right' for them. The enactment of this role, coupled with expectations of clinical authority, meant clinical decisions should not be questioned by the patient" [35]. The influence of health authority support was also sometimes highlighted as a positive influence, in contrast to technology manufacturers’ discourse: "Governmental endorsement […] was deemed crucial to acceptability. Congolese practitioners would not perceive evidence provided by other parties, such as manufacturers, as reliable" [46]. Peer influence also plays a role: "Some health workers are not regularly using the CDSS [clinical decision support system] during patient care, which demoralizes other providers who are using the system" [60]. Thus, the opinion of certain key individuals can have both a positive and negative influence on the acceptability of the technology. This influence depends on the opinion of these individuals or institutions on the technology, but also on the importance given to this opinion by the people whose acceptability is being measured.

Perceived advantages are the extent to which an innovation is perceived as bringing benefits. This is one of the dimensions most commonly found in the results of the studies included in the scoping review. Some of the perceived advantages include the improved diagnostic capacity and prescription of the right treatment [35], improved patient confidence in health workers [33], reduced consultation time [42], simplification of medical procedures [54], the reduction of unnecessary drug prescriptions [69], etc. All these benefits associated with the introduction of the new technology positively influence its acceptability and encourage its use and integration.

However, sometimes the new technology may seem to offer comparatively fewer benefits than what was previously used, which may have a negative impact on its acceptability. For example, some health workers pointed out that the RDTs for malaria (not able to quantify malaria parasites) were less accurate than the laboratory tests used in the past [33].

Compatibility is the extent to which an innovation is perceived as being consistent with the existing values, past experiences and needs of potential adopters. This dimension is, therefore, necessarily linked directly to the context. In some studies, the influence of the adequacy between the technology introduced and the setting of its introduction has indeed been highlighted: "the device is considered to have limited compatibility with equipment and competences of Congolese practitioners, especially of those active in rural settings" [46]. Here, the lack of compatibility between the technology introduced and the skills and case management possibilities influence the acceptability of the innovation negatively. Other elements may have an effect, such as its compatibility with the existing work organization in the health center. As pointed out in the article by Jensen et al. [38], in their case, the technology integrates well with the existing system: "eIMCI (electronic Integrated Management of Childhood Illness) was easy to accommodate within the daily staff allocation and clinic workflow". Compatibility can also be related to the perception of the tool itself: "Some of these interviewees said that needle length standards are defined for children of high-income countries" [37], or the adequacy with local values and customs, as underlined by this example of a technology using a blood test: " 'I need my husband to allow me to give the child's blood for testing.' […] 'my religion does not permit us to give blood.'" [33]. However, while this compatibility is a dimension that seems to be of important consideration, we found little referring to it in the articles.

We described the ‘context’ dimension as how the contextual environment (organizational, political, economic, social, etc.) influences the degree of acceptability of an innovation. While there was little mention of this dimension in the frameworks used to shape the a priori framework, we found many elements about it in the articles. As explained, we have included the context as a dimension that can impact the influence that each of the factors described above has on acceptability. For example, in a context where the health centers are highly attended, health workers "often noted that once the facility reached around 4 clients per SBA [skilled birth attendants], timely data entry was difficult", thus fueling the perceived disadvantages associated with the system [43]. Similarly, shortages of medicines and medical equipment [60], poor access to electricity in health centers [54], unclear national guidelines [33], or other elements related to the context can also feed the dimension on perceived disadvantages (respectively: impossibility of following the prescription recommendations provided by the innovation; difficulty in charging the device battery; confusion about procedures to follow, etc.).

Another element of the context that can influence acceptability, in particular via the "perceived complexity" dimension, is the level of familiarity with this type of technology or technologies in general. One article notes that the fact that point-of-care tests are regularly performed in health centers was positive because "All health workers agreed that the rapid test was easy to use due to the similarity with the HIV testing" [53]. Similarly, health workers who are familiar with mobile technologies may find a mobile application easier to use than others: "HCPs who did own an android phone demonstrated the ability to move through the application relatively quickly" [55]. On the contrary, unfamiliarity with technologies can have a negative influence, at least at the beginning: "Most HCPs were unfamiliar with touch-screen technology and thus were initially hesitant to use the device" [55].

Regarding perceived advantages, one study explains that in a setting where there is a "shortage of ophthalmic workforce", the new technology could help to overcome this lack of available staff by enabling task shifting and better human resource management [39].

Context can thus influence acceptability through all dimensions. These different examples clearly show the importance of studying the context in which the innovation is introduced and documenting the effects of the context on each of the determinants of acceptability.

Elements from the articles also led us to add the ‘intervention’ dimension (how an innovation is introduced) in our framework. Indeed, it is important to document and take into account how the intervention was designed, implemented, evaluated and by whom, as this can influence the acceptability of the innovation introduced. As previously discussed, this influence is also directly related to the context [81, 82].

For example, the type of person designated to use the innovation and the place where it is recommended to be used can influence its acceptability: "Using MAP [microarray patch] during outreach activities at fixed posts was also well anticipated by the majority of participants (average of 87.7%) […]. Using MAP at home during outreach strategies was perceived more cautiously (average of 60.2%)" [44]. Other elements related to the form of intervention, such as the role played by the monitoring team and the emphasis placed on the quality assurance program throughout the study [48], whether the intervention was included in a grant program [46], how the technology training was carried out and by whom [32], the quality of the instructions given both to health workers [54] and patients [54] before using the technology, the level of supervision of the intervention [53], etc., can all influence acceptability.

Therefore, how the intervention was shaped and implemented can influence acceptability both positively and negatively. It is suggested, however, that the high level of involvement of both health workers and the local population in the design and management of the program can enhance its acceptability: "They suggested that training the local population to run the program will overcome any potential obstacles related to acceptability and sustainability. Patients saw the importance of community participation as key to building trust and confidence in the program and put the population at ease" [39].