1. Introduction
Smallholder farmers in sub-Saharan Africa (SSA) face multifaceted and overlapping risks. For example, the simultaneous occurrence of malaria, animal diseases, pest infestations of crops, and climate change affect smallholder farmers [
1,
2,
3]. As smallholders, for whom agriculture forms the basis of their livelihoods, have a limited ability to cope with risks, any shock that reduces agricultural productivity can have significant impacts on their wellbeing [
4,
5,
6,
7,
8]. However, managing agricultural production constraints without alleviating human and animal health burdens might not generate significant and sustained benefits for achieving the desired development outcome (e.g., reducing hunger, malnutrition, and poverty). As such, building farmers’ resilience and adaptive capacity to co-existing production constraints and health burdens may require a holistic approach. This, in turn, calls for new and integrated approaches to address multiple constraints [
9,
10,
11,
12,
13,
14,
15,
16,
17,
18].
In this paper, we examine the impact of introducing integrated health interventions that address the co-existing constraints of farmers in Ethiopia. These interventions include an integrated vector management (IVM) package to control malaria, biconical (Ngu) tsetse fly traps to control trypanosomiasis, and push–pull technology to address stemborer and fodder shortages. All three interventions are eco-friendly, since no pesticides are used. This, in turn, mitigates biodiversity losses and public health problems associated with excessive use and misuse of pesticides. In addition to the above three interventions, improved beekeeping was promoted to relax farmers’ financial constraints by diversifying their income sources (see
Section 2.2 for details of the interventions).
Malaria, trypanosomiasis, and stemborers are major productivity-limiting factors in SSA. SSA accounts for more than three-quarters of the global cases and deaths due to malaria. Malaria accounts for 40% of the public expenditure in Africa [
19,
20]; as a result, malaria creates a huge economic burden on the region. Furthermore, livestock productivity and draught power performance are constrained by trypanosomiasis [
21,
22,
23]. Due to trypanosomiasis, nearly three million cattle die annually, resulting in a direct economic loss of 1–1.2 billion USD. It also kills more than 50,000 people annually, although the impact on humans is declining over time. Accounting for the indirect losses, SSA incurs up to 4.75 billion losses annually [
22,
24,
25,
26,
27]. Similarly, stemborer pests cause losses of up to 20%–88% in crop production in different parts of SSA, exacerbating food insecurity in the region [
28,
29,
30,
31,
32]. These problems are further aggravated by other biophysical and socioeconomic constraints, such as natural resource scarcity and degradation and poor access to markets [
4,
7,
8,
9].
Even though smallholder farmers encounter multiple constraints, development policies and agricultural interventions are often implemented in a stand-alone manner [
9]. Malaria control using insecticide-treated bed-nets and dichlorodiphenyltrichloroethane (DDT) improves human capital [
33,
34]. Positive gains were also documented from trypanosomiasis control using vaccines, insecticide-treated cattle, targets and traps, release of sterile male tsetse flies, and chemicals [
23,
35,
36,
37]. Furthermore, controlling stemborer infestations through parasitoids and push–pull technology was found to have economically significant gains [
38,
39,
40].
However, fragmented individual interventions addressing the human–plant–animal health challenges could be criticized for three reasons. Firstly, individual interventions may not be able to generate sufficient benefits to sustainably reduce poverty and food insecurity [
9,
10,
11,
18,
41]. Secondly, these interventions might have a small impact, or the impacts might not last long [
9,
12,
42,
43]. Thirdly, implementing individual interventions with little or no coordination could be expensive, which leads to the suboptimal allocation of scarce resources [
44,
45].
Cognizant of the above shortcomings, a small but growing body of literature has started to investigate the potential synergies and trade-offs among interventions. In terms of increasing productivity income and reducing risks, synergies among crop technologies have been documented [
46,
47,
48,
49]. Furthermore, addressing the constraints of the poor through multi-layered interventions could bring meaningful benefits compared to single interventions [
8,
12,
14]. Contrary to this literature, a few studies have empirically found that multiple interventions may not necessarily lead to higher benefits [
50,
51]. These groups of studies may imply that a big push through a simultaneous investment in many sectors may not be needed [
52]. The debate for or against multiple interventions seems to suggest that more research is required to fine-tune interventions that benefit resource-poor farmers.
The interventions and the empirical results reported in this paper offer one of the first attempts to address the multiple constraints of smallholder farmers. To our knowledge, no empirical study has so far reported on the impacts of integrated animal–human–plant health interventions. We also contribute to the mixed findings of previous studies on the economic gains of multiple non-health interventions in various contexts [
12,
14,
46,
47,
48,
49,
50,
51,
52]. The empirical results support the argument in favor of multiple interventions. Our results suggest that multiple technologies and strategies relaxing farmers’ production constraints and risks bring higher benefits than single interventions. Our findings fit into the theoretical literature that argues for a big push to address the social–ecological–institutional limitations that poor people face [
10,
17,
18,
53,
54]. Furthermore, the integrated approaches considered in this study are in line with the United Nations’ Agenda 2030, which calls for a holistic approach to achieving the Sustainable Development Goals. Thus, this paper provides timely evidence of the impact of the integrated interventions on rural households facing complex constraints.
The remainder of the paper is organized as follows.
Section 2 describes the study area and offers details on the individual interventions, as well as on the expected synergies among them.
Section 3 presents the data together with the whole-farm multiperiod linear programming model used to examine the economic implications of individual and combined interventions.
Section 4 presents our main results, while
Section 5 concludes the paper and points to its implications for future interventions and policy-makers.
5. Conclusions and Policy Implications
In this article, we provided evidence of the economic return of introducing multiple health interventions, which cut across animals, humans, and crops, into the existing farming systems in Ethiopia. We used a multiperiod linear mathematical programming technique in a whole-farm context. The approach we used enabled us to optimize farmers’ multiple objectives, including maximizing income, meeting household consumption, and livestock feed requirements. The interventions and the empirical results reported in this paper offers one of the first attempts to address multiple constraints of smallholder farmers simultaneously. The study provides the first empirical results on the impacts of integrated animal–human–plant health interventions. We also contribute to the mixed findings of previous studies on the economic gains of multiple non-health interventions undertaken in various farming systems.
Our findings indicate that significant income and resource productivity gains could be obtained from integrated animal–human–plant health interventions. Our study confirms that the simultaneous dissemination of integrated health interventions generates an income of 368 USD per capita higher than individual interventions. Furthermore, we documented that the income gains from multiple interventions are moderated partly through better resource-use efficiency and land productivity. Our results can contribute to building the evidence base for multiple interventions, which can provide valuable insight to policy-makers and development partners, for example, to revisit the extension system and to develop integrated interventions aligned with the Sustainable Development Goals to simultaneously tackle the numerous challenges facing smallholder farmers. However, the successful implementation of an integrated approach requires the cross-sectional coordination and building capacity of farmers, because to use more than one intervention can demand extra resources, which are often constraints for smallholder farmers.
While our study has demonstrated some interesting positive impacts of health interventions, the findings of this study may need to be interpreted with caution. First, the data for the estimation came from focus group discussions and cross-sectional household survey data, since baseline data were not collected. Additional evaluations using a well-designed experiment would allow a deeper understanding of the economics of integrated health approaches. Second, the model did not include risk and uncertainty explicitly, although household food consumption and livestock feed requirements were captured in the model. Third, the full benefits of the interventions have not yet been captured due to data limitations, which may underestimate their impacts. For instance, controlling tsetse flies and trypanosomiasis may improve livestock meat productivity. The interventions also have environmental benefits, but these benefits were not considered. Fourth, given that the case study was conducted in one district, there might be problems with inferences to the country level. Fifth, our model, like many other conventional representative farm models, fails to capture heterogeneity and interactions. Thus, bottom-up modeling approaches such as multi-agent systems could provide further insights for improved resource allocation and targeting by capturing heterogeneity and the complex interlinkages between the different production options, resource endowments, and heterogeneous responses to proposed policy interventions [
1,
2,
7,
8]. A final caveat is that, although we demonstrated the income gains of multiple interventions, we did not examine their relative cost-effectiveness due to a lack of accurate cost data. However, investigating the costs needed to achieve the income gains associated with each intervention, either individually or jointly, is critical for drawing a more robust recommendation on the allocation of scarce public resources [
72]. Future research that aims to assess and compare the relative cost-effectiveness of individual and integrated interventions is thus crucial.
Regardless of these limitations, this case study provides valuable insights into how integrated health interventions can have a substantial positive effect on improving farm households’ incomes. Development practitioners need to be encouraged to break through and link across the silos within their organizations. Embracing complexities instead of stand-alone silver bullets can support such endeavors toward identifying more synergistic interventions.