The annual cost of not breastfeeding in Indonesia: the economic burden of treating diarrhea and respiratory disease among children (< 24mo) due to not breastfeeding according to recommendation

Breastfeeding has been shown to provide many benefits to children, mothers, health systems and economies. Breastfeeding is associated with decreasing maternal risk of breast cancer, ovarian cancer and Type 2 diabetes, and higher intelligence of the child and decreased risks of infections, malocclusion of the teeth, overweight and diabetes [1, 2]. Breastfeeding according to recommendation consists of early initiation of breastfeeding (within the first hour of birth), exclusive breastfeeding for the first six months, and continued age appropriate breastfeeding at two years (accompanied by iron-rich complementary foods) [3].

Not breastfeeding according to recommendation is associated with diarrhea and pneumonia/respiratory disease (PRD) in children (< 24mo) [4‐8]. Studies show children (0-5mo) who were not breastfed have a 165% higher risk of suffering from diarrhea and 107% higher risk of pneumonia than children who were exclusively breastfed [5, 8]. Worldwide, not breastfeeding according to recommendation is attributed to the death of 823,000 children (< 5y) and 20,000 deaths due to breast cancer each year [1].

In Indonesia, 96% of children (< 24mo) are breastfed at least once, but only 42% of children (< 6mo) are exclusively breastfed [9]. As a result, not breastfeeding according to recommendation in Indonesia is estimated to contribute to 5377 preventable infant deaths due to diarrhea and PRD per year [10, 11].

Even though Indonesia has policy level laws and regulations supporting exclusive breastfeeding [12, 13], the ineffective monitoring, implementation, alongside other institutional barriers, prevents greater progress [14]. Wide scale issues such as aggressive marketing of breast milk substitutes (BMS) and limited workplace lactation spaces for breastfeeding mothers [15, 16] are still issues as a result of a lack of monitoring and enforcement of existing regulations. Although the World Health Organization (WHO) International Code of Marketing of Breast Milk Substitutes [12, 17‐21] has been accepted, it is not sufficiently enforced. Indeed, Indonesia is one of the largest contributors to the rapid increase of the growth of BMS sales in East Asia and low and middle income countries in general, only second to China [22]. The current Indonesian policy of 12 weeks of maternity leave [23] fails to meet minimum recommendation of 18 weeks set by the International Labor Organization [24]. The ideal period is six months, to enable mothers to exclusively breastfeed and adhere to WHO recommendations for exclusive breastfeeding [3].

The lack of accredited, “Baby-Friendly” breastfeeding promoting hospitals [25] and the lack of mass media support to facilitate a culture of breastfeeding [26] contribute to high levels of not breastfeeding according to recommendation. Moreover, given the diverse political and cultural setting of Indonesia, the country may benefit from a holistic approach to increase breastfeeding rates and Infant and Young Child Feeding Practices (IYCF) and among healthcare workers, doulas, religious figures, and village heads. Multi-sectoral intervention approaches can include advocacy, training, media promotion and home visits [27]. Some of these approaches have been ongoing in some areas of Indonesia and community-based training increases breastfeeding knowledge [27‐29].

Breastfeeding also has the potential to save costs for parents, insurance companies, employers, and society in general (e.g. healthcare costs, infant formula costs) [30‐32]. In seven countries in South East Asia, the annual treatment cost stemming from cases of childhood diseases attributed to not breastfeeding according to recommendation is large (US$293.55million), of which it was estimated that 87% of the costs came from Indonesia [10]. As Indonesia recently introduced national health insurance in 2014, these costs may be borne mostly by the government in the near future and this may put pressure on the already low health expenditure in Indonesia. The per capita healthcare expenditures adjusted for Purchase Power Parity (PPP) was US$3.02/capita in 2016 and was only slightly higher than that of India (US$2.69 PPP). Indonesia’s healthcare budget is the lowest GDP % among OECD countries at 2.8% (2016) [33].

The costs estimate attributed to not breastfeeding according to recommendation by Walters et al. [10] (including both provider and patient costs) are substantial. However, the healthcare system cost in the above-mentioned study was estimated using costs from one region in Indonesia and may not fully represent other regions in Indonesia. Therefore, the present study seeks to provide a more comprehensive national expenditure estimate by surveying five different regions in Indonesia.

This study addresses the following questions. First, using the sub-national level unit cost and utilization data for more accurate estimates, what are the health system costs of treating diarrhea and PRD for children not breastfed according to recommendation in Indonesia? Second, what is the out-of-pocket financial burden of patients’ family member(s) in accessing care?

We aim to present the economic impact of not breastfeeding according to recommendation, extrapolated by treatment costs of diarrhea and PRD among children (< 24mo) from the healthcare provider perspective and patient perspective. To our knowledge, such study in Indonesia is limited to one study [10]. Knowing the economic impact of not breastfeeding according to recommendation in Indonesia should spur policy makers and society in general into action to improve breastfeeding policies and practices.

The methods used were based on the Walters et al. and Bagriansky study [10, 34]. Provider costs were assessed in five different categories of healthcare facilities; public hospitals, private hospitals, village health posts (posyandu), primary health centres (puskesmas), and “others” being other healthcare facilities. For hospitals, cost data was collected separately for inpatient and outpatient visits. The “others” category consists of facilities not grouped within the other five types of facilities. For these, we used the treatment cost of the most similar of the five types of institutions from our survey data. The list of other facilities, is provided in an Additional file 1. Data collection from 2015 was used to estimate the costs from Bandung city, while in the rest of locations, data collection was from 2016. All costs were converted to USD using the 2016 reference exchange rate from Bank of Indonesia [35].

Data collection was performed in five separate sites, two cities and three non-urban districts, ranging from West to near East Indonesia. The five selected sites where; Bandung city (West Java), Serdang Bedagai district (North Sumatra), Tomohon city (North Sulawesi), Gianyar district (Bali), and Kupang district (East Nusa Tenggara). The site selection was based on discussions with stakeholders and experts from the Ministry of Health, the Ministry of Children and Women Empowerment, UNICEF Indonesia, and Alive & Thrive. The treatment costs obtained from each site were used to represent similar provinces. The list of provinces is provided in Additional file 2. In total, data was collected from 13 facilities.

The Gates Reference Case principles [36] was followed in conducting and reporting our study. However, since this study only focused on the cost of not breastfeeding according to recommendation, some principles, such as calculating Disability-Adjusted-Life-Years, was not pertinent. However, whenever relevant, the appropriate case principles were applied.

These results highlight the economic impact of not breastfeeding according to recommendation through the treatment costs of diarrhea and PRD in Indonesia. The annual healthcare system cost of not breastfeeding according to recommendation associated with associated with diarrhea and PRD amounts to US$119 million per year (1.6 trillion Rupiah), which is 0.01% of Indonesian Gross National Income (GNI) in 2012 [41]. Of this, 80% of the expenses comes from treating diarrhea (Table 2). From the provider perspective, the average treatment cost per case of both diarrhea and PRD is US$11.37, which is 10.6% of annual healthcare expenditure per capita of Indonesia in 2012 [42]. Translating these average treatment costs using 2016 purchase power parity figs. [43], we found that these treatment costs are lower than that of Laos, Thailand, and Viet Nam, roughly similar to Timor Leste, and around two times higher than that of Cambodia and Myanmar [10, 34, 44, 45]. Regardless, a significant share of these expenditures could be translated into cost-savings preventative measures by the Ministry of Health, and reallocated towards other priorities such as IYCF and breastfeeding promotion interventions. It is important to note, however, that the study by Walters et al. has shown that the highest non-medical costs of not breastfeeding according to recommendation in Indonesia came from the long-term impact of cognitive losses, amounting to US$1344 million per year. This was the largest impact when compared to the other countries in the study [10]. Moreover, the same study also showed that the amount of child and maternal death (from breast cancer) attributed to not breastfeeding according to recommendation was also the largest in Indonesia, amounting to more than 5000 and 2000 deaths respectively. Including these impacts to our current study would further highlight the egregious cost of not breastfeeding according to recommendation to the society, not to mention the further impact caused by the cognitive loss and death.

Secondly, the costs of treating diarrhea and PRD for children (< 24mo) amounted to approximately US$88 million from provider perspective (i.e. excluding patient perspective costs), and accounts for 75% of the total healthcare system costs. This leaves around 25% of the healthcare system costs to come from non-medical costs borne by patients (i.e. transport, productivity loss), amounting to around US$30 million per year (US$3.85 per case). Differentiating between diarrhea and PRD, we found that this percentage increased substantially for PRD (55%, US$6.00 per case), while the percentage is at 20% for diarrhea (US$3.33 per case). This shows that although overall the costs of diarrhea are higher than that of PRD due to higher number of cases, the latter cause much more burden to the patients through non-medical costs. Furthermore, PRD has a higher mortality rate in Indonesia [10]. Note that we did not put monetary value to caregivers for unpaid work at home. Therefore, the figures underestimated the financial burden on homemakers and stay-at-home fathers, and on providing outpatient care for the patients and palliative care. Therefore, these results undervalue the true productivity loss and, subsequently, the societal cost of diarrhea and PRD due to not breastfeeding according to recommendation. These costs may potentially be avoided if breastfeeding according to recommendation was done properly.

Thirdly, the patient out of pocket costs (including medical and non-medical costs) of accessing both outpatient and inpatient care for diarrhea and PRD treatment are a potential problem for families as they account for more than 10% of average monthly income [46], and much higher in the case of inpatient care. Costs requiring this much of the average income may cause patients to avoid visiting the facilities for treatment which may lead to more severe diarrhea and PRD cases, and may, to some extent, explains the almost 50% of untreated cases in our finding as it is a significant barrier to access treatment. In this context, providing breastfeeding according to recommendation to prevent diarrhea and PRD [4‐8] would be beneficial to patients’ family members as they can save financial resources. Be mindful that our productivity loss calculation may undervalue the true cost occurring as previously explained. In addition, not breastfeeding according to recommendation may necessitate purchasing BMS for infants which may further increase family expenses. Both breast milk substitute expenses and medical expenses due to diarrhea and PRD have been shown in other studies in the Philippines [47, 48]. Even though our study does not explore the cost of purchasing BMS and the subsequent medical expenses, these concerns should be addressed as it does affect the financial impact on families and may dictate whether they feel they can afford to seek medical attention or not.

In addition, a comprehensive breastfeeding strategy should be implemented to ensure breastfeeding is promoted, protected, and supported. The role of workplaces, healthcare workers, and social support from families cannot be underestimated. Creating an enabling environment for breastfeeding is crucial, and all stakeholders play significant roles. Providing longer maternity leave and supporting women to continue breastfeeding after returning to work are just two examples of how environments can be designed to support breastfeeding according to recommendation [49‐51]. Moreover, six months of maternity leave alone is insufficient to ensure exclusive breastfeeding if mothers do not have access to skilled counseling support and if social norms around breastfeeding are not supportive. The World Bank estimated that scaling up a core comprehensive breastfeeding strategy to achieve the Global Nutrition Target for breastfeeding could generate a return on investment of US$35 dollars for every US$1 invested across all low-and middle-income countries [11]. With its growing economy and demand for a skilled workforce, Indonesia can benefit economically from a similar investment in interventions to protect, promote, and support breastfeeding.

In terms of study limitations, Indonesia is an extremely diverse nation, and a broader sampling of treatment costs is needed for more accurate results. Data regarding treatment costs were assessed from two cities and three districts to represented the treatment cost of 98 cities, 416 districts and 34 provinces in Indonesia. Furthermore, our study sample to calculate patient costs is relatively small (615 respondents) in the context of the number of people in Indonesia (more than 250 million people). However, we found that patients per type of facility in each of the city/districts have similar trends in terms of distance/time travelled, and monthly income/expenditure. Therefore, as our selection of cities/districts is spread in five different parts of Indonesia, we believe that our sample, to some extent, may represent the trend in Indonesia. Future studies can explore treatment and patient costs from additional regions, so more islands or regions in Indonesia will be represented by the extrapolations of data gathered here.

Another limitation was the assumption that the relative risk of diarrhea and PRD due to not breastfeeding according to recommendation was evenly distributed across Indonesia [10]. Further study on calculating this relative risk specifically for each region in Indonesia is required, as there is a high possibility that variation is high. This, in turn, may yield better cost estimation.

Lastly, the productivity value of housewives and stay-at-home dads were not estimated for the productivity loss and extra expenses due to providing outpatient care. Therefore, to some extent, this undervalues our calculation of non-medical cost. Future study may include these non-medical costs, and may go even further to estimate the costs of palliative care and future productivity loss of children and maternal breast cancer deaths attributable to not breastfeeding according to recommendation and provide a more complete picture of the severe impact of not breastfeeding according to recommendation in Indonesia.

Although our study is based on the methodology used by Walters et al. [10], our study results differ as the mean treatment cost for both diarrhea and PRD combined was US$15.22 versus US$21.10 as found by Walters et al. One of the main reasons for this was the Walters et al. assessed treatment cost from only Bandung city, whereas in this study we included data from one additional city and three districts in different regions to represent the costs from different provinces. Provincial figures of different diarrhea and PRD prevalence was also used. Our study found that costs were generally higher in Bandung city than other regions. Therefore, our results may better represent the cost differences throughout Indonesia, and includes a greater emphasis on lower cost regions/locations.

The 2016 annual economic impact of not breastfeeding according to recommendation in Indonesia based on the high cost of treating diarrhea and PRD warrants urgent attention. The healthcare system costs and non-medical out of pocket cost were US$11.37 and US$3.85/treatment respectively. The total healthcare system costs were US$118.62 million annually. Non-medical costs from patient perspective accounted for 25% of treatment costs. Out of pocket costs represented a significant burden and a financial barrier to seeking treatment as it accounts for more than 10% of the caregiver’s average monthly income.

Importantly, these figures only include immediate costs stemming from not breastfeeding according to recommendation and do not take into account the cost of future cognitive losses, which could reach US$1.54 billion per year for Indonesia [10].