Background

The global healthcare worker (HCW) shortage disproportionately impacts low- and middle-income countries (LMICs) [1]. This has been a particular challenge for the HIV response in LMICs, such as South Africa [2,3,4,5]. South Africa has the world’s largest HIV epidemic [6], which requires sufficient human resources for health to ensure HIV programs can be fully delivered [5, 7]. Insufficient human resources lead to excessive workloads for HCWs, which is known to reduce HCW motivation and contribute to attrition [8,9,10]. Use of lay health workers (LHWs), defined as those supporting health work without professional or paraprofessional qualifications, has long been advocated to increase human resources for health and thereby improve health care [11, 12]. LHWs are a recommended approach to support HIV service delivery [12, 13], as LHWs can assume HIV testing and counseling responsibilities to increase testing, treatment initiation, and retention in care [13,14,15,16,17]. Novel strategies are need to bolster the supply of LHWs, which remains insufficient in LMICs, including South Africa, due to funding constraints [18,19,20,21].

Youth Health Africa (YHA) is a novel LHW approach that places young adults needing work experience in one-year, non-clinical internships at health facilities to strengthen HIV service delivery. Like traditional LHWs, interns support clinics by task shifting and task sharing [22]. However, YHA is different from other LHW programs as it involves young adults working for only one year, when typically, LHWs in South Africa are older with longer assignments [20]. Moreover, YHA is a multisectoral program; while YHA seeks to strengthen the health system through LHWs, its primary goal is youth empowerment, providing interns work experience to increase youth employability and decrease unemployment [23].

Research suggests YHA may improve facility-based HIV service delivery [24], but HCW perspectives on this program are unknown. HCWs must perceive the YHA approach as acceptable and appropriate for it to be recommended as a LHW approach, as acceptability and appropriateness are paramount to successful program implementation [25, 26]. While HCWs are often appreciative of LHW programs because they result in task shifting and task sharing [27], HCWs can find LHW programs less acceptable or appropriate due to LHW incompetence or increased workplace competition [3, 11, 26, 28]. The novelty of the YHA approach means it is difficult to predict how HCWs perceive the program. We thus conducted a study to [1] describe the acceptability and appropriateness of YHA from the HCW perspective and [2] understand what promotes or hinders high acceptability and perceived appropriateness of the YHA approach.

Methods

Study design

We used a convergent mixed methods study design (QUAN + QUAL) to assess HCW perceptions of acceptability and appropriateness of YHA as implemented in South Africa. We defined acceptability as how satisfying or agreeable YHA was to HCWs and appropriateness as how well YHA fit HCW/facility needs [25]. We concurrently collected quantitative data through surveys and qualitative data through interviews using a multilevel sample of HCWs. We analyzed data separately, then merged findings for interpretation. We used a mixed methods approach for purposes of complementarity and triangulation, which fostered a more comprehensive understanding of HCW perceptions than either method would have alone [29].

YHA program

YHA collaborates with PEPFAR implementing partners who support clinics to place interns in facilities that need additional human resources to support HIV service delivery. YHA interns are 18–34 years old with secondary education but no employment experience. They are assigned to be program interns (e.g., HIV testers, peer navigators) or administrative interns (e.g., data entry, filing clerks). YHA leads a broad 3–5-day training for interns, with health facilities providing technical training as needed. YHA pays interns stipends and provides monthly professional development trainings. Interns work alongside HCWs at facilities and are formally supervised by a HCW working for the PEPFAR implementing partner. YHA is modeled after internship programs implemented at South African businesses, where private organizations fund internships, incentivized by the country’s Broad-Based Black Economic Empowerment (B-BBEE) policy [30].

Study population

The study population comprised HCWs who supervised or worked with YHA interns, which we refer to as “intern supervisors” and “frontline HCWs”, respectively (Supplement 1). Intern supervisors were tasked by the PEFPAR implementing partner to provide supportive supervision to interns and often worked across many facilities. Frontline HCWs were staff who worked alongside interns (co-workers) and the nurses who helped manage/lead the HIV unit (facility managers).

We conducted surveys with intern supervisors because they offered a broad perspective on YHA, having engaged with multiple interns across many facilities. We interviewed frontline HCWs to gain in-depth understanding of their experiences working with interns and perceptions of program benefits.

Setting

While YHA has been implemented across South Africa, this study was conducted among HCWs at facilities in Gauteng and North West (NW) provinces where Aurum Institute is PEPFAR implementing partner. Surveys were completed by intern supervisors working across these provinces. Interviews were conducted in facilities from two districts: Ekurhuleni (Gauteng) and Ngaka Modiri Molema (NW).

Sampling

All intern supervisors supporting YHA were invited to take the survey. Concurrently, a sample of frontline HCWs were selected for interviews using a two-step process. First, we selected facilities using random sampling, stratifying by geography (NW versus Gauteng) and facility type (clinic versus community health centers [CHCs]/hospitals), as we hypothesized that experience with YHA may vary based on these elements. Five clinics and three CHCs/hospitals were selected from each region. We then used purposive sampling to select frontline HCWs to interview. At each facility, we selected an available facility manager and a co-worker who worked closely with a YHA intern.

Data collection

From March–May 2021, intern supervisors completed a self-administered, electronic survey, which included close-ended questions: 15 to measure acceptability, 15 to measure appropriateness, and others on supervisor role, supervisor characteristics, and program implementation. The questions on acceptability and appropriateness were adapted from the Johns Hopkins Dissemination and Implementation Science (JHD&I) Measure of Acceptability and Appropriateness [31], which was validated and used in other LMICs [32,33,34,35,36]. These questions were answered on a four-point scale (1 = “Not at all,” 2 = “A little”, 3 = “A moderate amount,” or 4 = “A lot”). We made minor adaptions to fit YHA and South African contexts and piloted the survey before use. The survey was in English and implemented using REDCap electronic data capture tools, which were hosted at the Institute of Translational Health Sciences.

During the same period, we conducted semi-structured in-person interviews with frontline HCWs, which included questions on their experience with YHA and benefits and challenges of the program. Two questions asked about acceptability and appropriateness, pulled from the JHD&I measures [31]. Research assistants from Aurum Institute conducted interviews in English or local languages (i.e., Setswana in North West and IsiZulu or Sesotho in Gauteng), based on interviewee preference. Interviews lasted between 30–60 min. All interviews were recorded, transcribed and translated into English.

Quantitative analysis

We calculated a total and average score for acceptability and appropriateness for each respondent. When responses were missing or categorized as “I don’t know”, we weighted final scores based on existing responses. We then calculated the mean and standard deviation for acceptability and appropriateness in Stata 15 (StataCorp LLC., College Station, TX). We graphically summarized responses for each question comprising the acceptability and appropriateness measures in R 3.6.1 (R Core Team 2019, Vienna, Austria), using ggplot2 and HH packages [37, 38]; missing responses or those responding “I don’t know” were excluded. Finally, we conducted bivariate logistic regression analyses in R to explore whether intern supervisor characteristics or their roles were associated with acceptability and appropriateness.

Qualitative analysis

We used Dedoose 9.0.17 (SocioCultural Research Consultants LLC., Los Angeles, CA) to conduct an inductive, thematic analysis. A primary coder developed a codebook based on preliminary review of transcripts. The primary and secondary coder coded three transcripts independently and came to consensus on code applications and an expanded codebook. The primary coder coded all transcripts and further refined the codebook. The secondary coder reviewed all coded transcripts and the updated codebook. Both came to consensus on all application of codes and collaboratively identified themes emerging from the interviews.

Integration

We compared quantitative and qualitative findings and interpreted results together to present a holistic view of acceptability and appropriateness from the HCW perspective. To facilitate integration, we matched survey questions and interview themes to the domains of the Theoretical Framework of Acceptability, which was created to facilitate an understanding of what impacts acceptability of healthcare interventions [39]. While this framework’s definition of acceptability is inclusive of appropriateness, we added a domain, ‘facility fit’, to better fit Proctor’s definition of appropriateness [25]. To further support interpretation of our data, we divided this framework’s domains into three naturally occurring groups: nature of the intervention, implementation of the intervention, and results of the intervention. The domains of this framework are outlined and defined in Fig. 1.

Fig. 1
figure 1

Overview of the expanded Theoretical Framework of Acceptability (TFA) domains used in mixed methods analysis. Original domains are as defined by Sekhon, et al. [39], with addition of one domain, ‘facility fit’, influenced by the definition of ‘appropriate’ by Proctor, et al. [25]. Domains were grouped into three groups by this study’s authors

Full size image

Ethics, consent, and permissions

This study was approved by University of Witwatersrand (Johannesburg, South Africa) and University of Washington (Seattle, USA), along with provincial and district-level ethics committees where it was conducted. Participants provided written consent before engaging in surveys or interviews.

Results

Participants

Sixty of 66 eligible intern supervisors completed surveys (91% response rate) (Table 1). Altogether, they supervised 364 interns at 180 facilities. Almost half had served as intern supervisor for at least one year, giving them experience with more than one intern cycle. We also interviewed 33 healthcare workers (16 facility managers, 17 co-workers) from 16 facilities (Table 2). Over half of facilities hosted 2–3 interns at the time when interviewed. Almost half of interviewees had worked at their facility for at least three years.

Table 1 Characteristics of survey respondents (n = 60)
Full size table
Table 2 Characteristics of interviewees (n = 33) and their facilities (n = 16)
Full size table

Quantitative findings

Surveys found YHA was acceptable and appropriate to intern supervisors. On a four-point scale, the average score for acceptability was 3.5 (standard deviation: 0.4), while the average score for appropriateness was 3.6 (0.4). Fifty-three respondents (89%) had average scores that rated the program at least moderately acceptable (≥ 3.0), and 57 (95%) had average scores that rated the program at least moderately appropriate (≥ 3.0) (Fig. 2). There were no associations with respondent characteristics (gender, profession, years in current position) or their roles with YHA (duration as supervisor, number of interns supervised, number of facilities requiring supervision) and acceptability or appropriateness (Supplement 2). Intern supervisors in Gauteng and those supervising only administrative interns had lower perceptions of acceptability and appropriateness, but differences were not significant.

Fig. 2
figure 2

Distribution of overall average acceptability and appropriateness scores for intern supervisors, by province (n = 60). Average scores of acceptability and appropriateness correspond to the following scale: 1 = Not at all, 2 = A little bit, 3 = A moderate amount, or 4 = A lot

Full size image

While acceptability and appropriateness measures were highly scored overall, there was a range in positive responses for individual questions (Fig. 3). For the acceptability measure, top scoring questions were related to the nature of the program and its results, while top scoring questions for the appropriateness measure were related to program results and its general fit with the health facility. For both measures, the lowest scoring questions were related to the clarity and burden of the intervention and competing priorities. Two questions related to satisfaction with program orientation and support provided for supervision scored much lower than other questions (Fig. 3A).

Fig. 3
figure 3

Summary of responses to acceptability and appropriateness questions on survey (n = 60). A: Responses to questions from perceived acceptability measure. B: Responses to questions from perceived appropriateness measure

Full size image

Qualitative findings

Most frontline HCWs reported that YHA was highly acceptable and appropriate. No respondents reported low levels of satisfaction or fit with YHA. Qualitatively, we observed no patterns in perceptions of acceptability or appropriateness by HCW role (facility managers versus co-workers), facility type, or geography. One nurse’s words capture the majority opinion:

“There are no disadvantages. I have been singing praises thus far, that they [the interns] have helped us a lot.” (NW-14, Clinic)

We identified four major themes that explained HCW perceptions on acceptability and appropriateness of YHA (Table 3).

Table 3 Key themes from interviews with frontline healthcare workers (HCWs) (n = 33) with supporting quotations
Full size table

Theme 1. YHA is mutually beneficial to HCWs and interns

Frontline HCWs expressed strong affinity for YHA because they viewed it to be mutually beneficial, benefiting HCWs by being “extra hands” to do work at facilities and benefiting interns and broader society by providing youth with employment and job training. This theme is a synthesis of two key ideas present in interviews: that interns reduced HCW burdens and that HCWs felt good helping interns.

HCWs strongly emphasized that they were overworked as facilities were understaffed. They reported that interns alleviated some of this burden by offloading work through task sharing with non-clinical staff and task shifting with nurses.

“Knowing that every day there is a person who is going to help with the simple tasks reduces the workload so that we can focus more on the clients.” (Gauteng-4, Nurse, Clinic)

In describing how YHA reduced facility workload, HCWs emphasized the interns’ professionalism. HCWs frequently noted that interns were young and lacked experience, but they described interns positively (e.g., respectful, hardworking, committed, and amiable). While interns did need initial training and occasional coaching on professionalism, they were ultimately viewed as competent, independent workers who took initiative to accomplish tasks.

“They are able to do everything by themselves, without us asking them, "Did you do this? Did you do that?" They know exactly what to do.” (NW-6, Nurse, CHC/Hospital).

While HCWs liked the YHA program because it helped them, it was common for HCWs to say the first reason they liked the program was because it benefited interns. HCWs described appreciating that YHA provided immediate support to youth (e.g., wages to support their families), how it empowered youth to have better futures, and how it would better society by improving the skills of South Africa’s workforce. Co-workers often described liking the program because it provided immediate relief to a youth’s challenging economic circumstances, while facility managers, who tended to be older, found satisfaction in mentoring youth, as they believed this mentorship would improve the youth’s, and society’s, future.

“I like [YHA] because at least they give these kids a chance of working. You know, some of them are coming… from schools, some of them are mothers, which they don’t have a job, they need a job you know.” (Gauteng-9, Data capturer, Clinic).

“[YHA] has also given me the satisfaction as a human being….that at least I am contributing to this, to our [youth]…., who will be our future tomorrow.” (Gauteng-13, Nurse, CHC/Hospital)

Theme 2. Facilities need agency when implementing YHA

While HCWs were generally enthusiastic about YHA, they expressed frustration about being excluded from the planning process. HCWs described feeling unprepared for interns and felt interns were sometimes assigned to roles that did not match facility needs because HCWs were not consulted.

“We only got clerks….Our challenge here is we don’t have data capturers…I wish they could have brought more data capturers and counselors.” (NW-3, Nurse, Clinic)

However, once interns were placed at clinics and HCWs assumed responsibility for the interns, they integrated easily into clinic operations.

“They have blended simply, and they have managed to just come to the facility and become part and parcel of us and became like a family.” (NW-14, Nurse, Clinic)

Having this agency promoted the sense that YHA fit the facility, which fostered a high affinity for the program among most HCWs. Facility managers experiencing the poorest communication with YHA were apt to feel they lacked agency in directing the intern’s work. Though outliers, these HCWs expressed less satisfaction with the program.

Theme 3: Short internships can be burdensome

HCWs explained that while interns were helpful, the short internship period could be burdensome. HCWs were required to provide interns on-the-job training, which was time consuming, specifically for co-workers working closest with interns. Many HCWs were concerned about this upfront investment as interns would be present for only one year. Moreover, HCWs expressed concern that they had come to depend on interns, who were on time-limited contracts. HCWs frequently described wanting to lengthen internships or find a way to permanently hire interns to reduce this burden.

“It is difficult when you have people you have trained then they have to go, then you have to start again. And we are not getting any remuneration for that, so it is tiring.” (Gauteng-12, Nurse, Clinic)

Theme 4: Success from YHA is found in more than numbers

While frontline HCWs believed YHA helped them reach HIV targets, they appreciated YHA because it brought success in other ways. HCWs described how interns created a “smoother” working environment, such as improving patient flow by expediting retrieval of patient files. HCWs also described how YHA improved morale, as it reduced workload and ensured staff could step away without worry of interruption to services.

“I feel better because I am no longer overloaded with work. Even if I am sick, I know there is a back-up then. The [YHA] data capturer will come and work. Everything will be well. (Gauteng-11, Data Capturer, CHC/Hospital)

HCWs also believed YHA benefited patient care, especially in terms of improved services for young people. Many HCWs explained their struggles to test young people for HIV, due to limited hours for HIV testing and counseling and large age differences between counselors and adolescents. HCWs felt YHA interns bridged these gaps by enabling facilities to provide testing for extended hours with younger counselors.

Integrated findings

Both intern supervisors and frontline HCWs found the YHA approach to be highly acceptable and appropriate. When we integrated findings from these two groups using the Theoretical Framework of Acceptability (TFA), we saw that intern supervisors and HCWs expressed similar sentiments about the YHA approach (i.e., the quantitative and qualitative findings converged), even though these HCWs had different roles in the program. This enabled us to identify elements of the program and its implementation that promoted or hindered acceptability and appropriateness of YHA among HCWs, which we discuss by TFA domain (Table 4).

Table 4 Joint display of factors promoting or hindering acceptability and appropriateness of Youth Health Africa
Full size table

HCWs genuinely liked YHA as they found it provided the facilities with much-needed support (TFA domain: affective attitude) and they agreed with its higher goal of empowering youth through the internships (ethicality). Other elements that promoted acceptability/appropriateness of the program included HCWs finding it easy to integrate interns into their jobs (self-efficacy), the program aligning with facility culture (facility fit), and HCW belief that interns helped facilities be more successful (perceived effectiveness).

Elements hindering acceptability and appropriateness included having an inadequate introduction to or communication with the program (intervention coherence) and the time required to train, and to a lesser extent, supervise interns (burden). We noticed divergence in whether HCWs perceived the opportunity cost of YHA interns to promote or hinder acceptability/appropriateness; while the benefits of YHA were widely believed to outweigh costs, frontline HCWs worried this balance could reverse if there was not a plan to ensure a sustained intern presence at facilities.

Discussion

In this study we explored HCW views on acceptability and appropriateness of a novel strategy to increase human resources for health: placing youth interns as temporary LHWs in health facilities to support HIV services. HCWs found this to be a highly acceptable and appropriate strategy to strengthen HIV service delivery because the nature of the program was appealing, specifically its altruistic goals; implementation of the program was easy, once interns were placed at facilities; and the program yielded positive results. However, aspects of program implementation, namely the upfront burden posed by interns and limited communication with HCWs, detracted from the program’s overall acceptability and appropriateness. Overall, HCWs expressed strong affinity for YHA because they believed its benefits—to HCWs, patients, and interns—outweighed its costs.

Our results align with past research that found HCWs generally have favorable views of LHW programs that support HIV services [27], but the reason for the strong acceptance of the program was unique. The predominant finding in our study was that HCWs liked the YHA program because it was mutually beneficial. Similar to past research, HCWs appreciated the program because LHWs benefited the facility in important ways not captured by standard health metrics (e.g., reducing workloads and improving clinic functions) [11]. HCWs described overwhelmingly positive interactions with interns and did not describe worries about competition or incompetence found in other LHW programs [11, 26, 28, 40,41,42]. Most interestingly, YHA’s goal to empower youth, which is unique to this LHW program, appealed to HCWs’ sense of altruism, which strongly contributed to their favorable perception of the program. Altogether, these findings suggest that HCWs find the YHA program to be as acceptable and appropriate as other LHW programs, if not more so, precisely because of what makes YHA different from other LHW programs: its altruistic focus on empowering youth.

Despite the overwhelmingly positive reception of YHA, its implementation can be improved to maximize acceptability and appropriateness as it is scaled. Our findings emphasized that facility HCWs need to be included in planning for interns. This is echoed in past research, which shows HCWs are, unsurprisingly, happier with LHW programs when they have a voice in how the program is implemented, specifically with regards to how LHWs are introduced to facilities [27, 28]. Similarly, it is imperative that program implementers consider extending the length of the internship period to offset the upfront burden of training interns; otherwise, the costs of the program may eventually outweigh its perceived benefits.

Policy implications

YHA’s appeal to HCW altruism through the program’s focus on youth empowerment may make this program attractive to policy makers striving to find new ways to strengthen human resources for health. Altruism is a key driver of HCW happiness and motivation, which has potential public health impact, as it can lead to reduced HCW attrition [9, 10]. To support HCW retention, researchers have proposed creating opportunities for HCWs to mentor others, as mentoring can improve morale by appealing to HCW altruism [10]. Our study suggests YHA could be such an opportunity, as it appears to improve HCW morale by enabling HCWs to engage in altruistic mentoring, while providing extra hands to reduce workloads, and thus lead to better workplace environment [8].

Moreover, the program’s multisectoral nature could make it attractive to policy makers as it could open new funding options for LHW programs. YHA has been funded through B-BBEE contributions, an economic empowerment policy [30]. The economic development sector, such as the African Development Bank, supports youth employment projects and could thus be a potential funder [43, 44]. The South African government has expressed interest in YHA because of its youth employment focus, while YHA’s focus on health system strengthening and HIV could also make it of interest to traditional public health entities (e.g., Department of Health and PEPFAR). A stakeholder analysis could be useful to elucidate their perspectives.

Limitations

Although this mixed methods study provides a robust assessment of HCW perceptions of YHA, it was subject to several limitations. Firstly, we could only access HCWs working at facilities associated with Arum Institute and did not include facilities supported by other implementers of YHA; this limits the generalizability of our results, but we believe key drivers of acceptability and appropriateness would remain similar across facilities. Secondly, the validity of our survey questions is uncertain in our context, as key questions were validated in LMICs other than South Africa, and their original use was for mental health programs [36]. Thirdly, our interviewers were employees of Aurum, which helped implement YHA. This affiliation may have led to social desirability bias, with feedback perceived as impacting future Aurum support. However, similarity in findings between surveys and interviews suggest challenges with survey validity and social desirability bias were minimal. Fourthly, our study was limited to HCWs and did not address the experience of interns or patients. Finally, while our research highlights how HCWs perceive YHA to impact health facilities, we did not objectively measure such impact (e.g., morale, patient wait time). Nonetheless, this study is important as it provides a robust assessment of how HCWs perceive YHA when implemented under routine conditions.

Conclusion

HCWs perceived having youth interns as temporary LHWs in facilities to be an acceptable and appropriate strategy to support HIV services in South Africa. While aspects of YHA implementation could be improved, HCWs found the program overwhelmingly acceptable and appropriate because its goals resonated with them, and they believed the program benefitted HCWs, patients, and interns. In particular, HCWs appreciated YHA’s altruistic nature, finding satisfaction in mentoring youth, suggesting this may be an especially effective LHW approach to strengthen human resources for health. Moving forward, we recommend adjustments to YHA’s implementation strategy to reduce burden on HCWs and research to quantify the program’s total cost and impact on patients and interns.