Garg et al. provide a useful, but somewhat incomplete, economic perspective on Community Management of Acute Malnutrition (CMAM) in India. Potential returns, like cost per recovered child, adjusted for spontaneous improvement under the existing system, are crucial for policy makers. Sixteen weeks after completion of the treatment phase (sustenance phase), only 123/838 children (14.7%) met the definition of recovery.1 For simplicity’s sake, we ignore: (i) anticipated lower recovery rates in public programme settings; and (ii) costs for linkages with the government-run Anganwadi centres for supplementary food during the sustenance phase.1 With these assumptions, our calculation of costs per recovered child are 6.8 (100/14.7) times higher than those of Garg et al.: US$ 1575/- (Rs. 97,650/-) and US$ 381 (Rs. 23,622/-) in research and Government settings, respectively. Further, annual budgetary requirements may be considerably higher due to non-response, relapse, and fresh cases of Severe Acute Malnutrition (SAM). Data from rural Meerut, near Delhi, provides a ballpark estimate of spontaneous recovery rates (27%) within a similar follow-up period.2 Actual costs per child recovered, even unadjusted for potentially equivalent spontaneous healing, are thus much higher than those quoted by Garg et al. However, they offer no cost-effectiveness analysis reporting cost per life saved or Disability-Adjusted Life Year averted to enable robust comparisons with the existing system or other interventions.

The cost of A-HPF (Augmented, Energy-Dense, Home-Prepared Food) was overestimated: (i) food was supplied at 1.5 times assumed requirements along with fuel, which merited costing or additional benefits adjustment; (ii) in Government programme, food procurement costs will be substantially lower because of expected integration with Public Distribution System and National Food Security Act; (iii) for bulk procurement, multivitamin mineral mix costs would be lower than pharmaceutical syrup used; (iv) as with exclusion of hospitalisation expenses, iron supplementation costs for anemic children should not be included because it is part of the National Iron Plus Initiative; (v) long-term benefits for the index child and family members, of extra counselling time for preparing various recipes needed to be factored in. Similarly, in the RUTF-L (Locally produced ready-to-use food) arm, employment and local economy benefits could have been accounted for.

It is unrealistic to expect a meagrely incentivised (INR 150) ASHA with several competing responsibilities to provide the same intensity and efficiency of ‘peer support’ as in the trial (incentive INR 700). The government guidelines cited to justify the ASHA incentive actually pertain to USHAs in urban contexts, which are awaiting implementation. Considering the potential for spontaneous recovery and lack of robust cost-effectiveness data from a “real world” Indian setting, a recommendation for scaling up CMAM through ASHAs is very premature.

Finite resources for public health spending invariably compel governments in LMICs to choose among interventions labelled as “very cost-effective”. An illustrative dilemma in Government setting comprises choices between cost of recovery for one SAM child, maternity cash entitlement (INR 6000) for four women, enrolling 20-24 additional families for Ayushman Bharat (annual health insurance cover of INR 5 lakh)3, or eggs twice a week for 46-58 children. All stakeholders, particularly the intended beneficiaries, should be an integral part of such selection process.4

References
1. Bhandari N, Mohan SB, Bose A, Iyengar SD, Taneja S, Mazumder S, Pricilla RA, Iyengar K, Sachdev HS, Mohan VR, Suhalka V, Yoshida S, Martines J, Bahl R, for the Study Group. Efficacy of three feeding regimens for home-based management of children with uncomplicated severe acute malnutrition: a randomised trial in India. BMJ Global Health 2016;1:e000144. doi:10.1136/bmjgh-2016-000144.
2. Sachdev HS, Sinha S, Sareen N, Pandey RM, Kapil U. Survival and recovery in severely wasted under-five children without community management of acute malnutrition programme. Indian Pediatr 2017;54:817-24.
3. FE Bureau. Modicare: Cabinet nod for Ayushman Bharat. Financial Express, New Delhi, March 22, 2018. https://www.financialexpress.com/industry/modicare-cabinet-nod-for-ayush... accessed April 18, 2018.
4. Marseille E, Larson B, Kazi DS, Kahn JG, Rosen S. Thresholds for the cost–effectiveness of interventions: alternative approaches. WHO Bulletin 2015;93:118-24.

Donkin et al have highlighted the constructive steps being taken to implement policy change facilitating the Social Determinants of Health (SDH) across the globe 1. Although progress is not universal, what has been achieved deserves praise.

We write, however, to highlight one key omission from the standard SDH model: religious faith. This deserves greater recognition as a social determinant of health for two reasons. First, is scale: a recent study demonstrated that 84% of the world’s 7.4 billion people affiliated themselves to a religious group 2. Second is the impact of religious faith on health, shaping both health beliefs and use of healthcare services 3.

Theories of supernatural causation of illness are ancient and diverse. They are also universal: a 1980 study of health belief systems worldwide found evidence that supernatural causes of illness “far outweigh” natural ones 4. Of course many such models may be counter to the Western biomedical model. They should, however, still be acknowledged, not least because when believers encounter Western biomedicine the two models typically become mixed without any sense of conflict.

The close link between healthcare services and religion is evident in both the historical (for example, medieval Christian hospitals in Europe) and contemporary (e.g. faith-based non-governmental organisations worldwide) contexts. The relationship between religion and healthcare is not confined to simple delivery. Religious faith can influence the way individuals approach and access healthcare services available 3.

If evidence of the importance of religious faith is required, one need look no further than social and psychological resilience. Religious faith features powerfully in the stories of some of the most vulnerable populations, for example survivors of military conflict and torture. Faith is also recognised as an independent protective factor against depression and suicide 5. Increased rates of migration, conflict, and the epidemiological transition towards chronic physical and mental conditions are likely to make religious faith more relevant to health on a national and global level.

We argue that faith operates beyond social capital and may not be captured through other secular social concepts such as community, social cohesion or culture. Researchers and policymakers are unlikely to harness positive aspects of religious faith if its status as a key social determinant of health is unacknowledged. Such recognition is also important to ensure that the influence of religious faith – positive and negative – is incorporated in undergraduate and postgraduate medical curriculums.

We hope that this letter serves as a stimulus to greater recognition of religious faith within the social determinants of health.

References:

1. Donkin A, Goldblatt P, Allen J, Nathanson V, Marmot M. Global action on the social determinants of health. BMJ Glob Health. 2018;3(Suppl 1):e000603.

2. The Changing Global Religious Landscape. Pew Research Center; April 5, 2017.

3. Shenouda JEA, Cooper MJF. "One Big Family": Pastoral Care and Treatment Seeking in an Egyptian Coptic Church in England. J Relig Health. 2017;56(4):1450-9.

4. Murdock, GP. “Theories of Illness. A world survey”. University of Pitsburgh Press, 1980 page 26.

5. Norko MA, Freeman D, Phillips J, Hunter W, Lewis R, Viswanathan R. Can Religion Protect Against Suicide? J Nerv Ment Dis. 2017;205(1):9-14.

I read the article “Implementing One Health as an integrated approach to health in Rwanda” by Nyatanyi et al1 with great interest and wish to share our efforts on Implementation of One Health in a Rural Medical School following concept of Public Money Stewardship. We have been working on human health coupled with ecological studies for a long time; however, we started working on One Health after sanction of a project on zoonosis by a joint task force of Indian Council for Medical Research and Indian Council for Agricultural Research (ICMR-ICAR) in 2015. The work is being carried out in project mode but we are working on various factors, as given below, so that we may give guidelines to the government for its implementation in programme mode, now preferably in a manner similar to that given in the study by Nyatanyi et al1.

One health laboratory: We have created a facility harbouring a containment laboratory, a BSL-2 laboratory as well as separate laboratories for serology and molecular testing. Planning for this facility included assessment of workload and availability of space and funds. Since we had limited funds made available to us from public money, we designed the facility for optimum use and thus followed the concept of public money stewardship given by Mahatma Gandhi whose principles we follow in our Institute. We also wanted to make this laboratory to be environmentally conscious and thus used solar panels to power it in a hybrid manner.2 Total fund utilized for renovation and equipment procurement were USD 170,000 and we can work on all the organisms which can be handled in a BSL-3 laboratory including viral and Orientia tsutsugamushi culture. We are also capable of performing PCR and RT PCR. We now propose that such One Health laboratories may be established at district headquarters.

Integrated diagnostic platform: Our laboratory is unique in the manner that we use integrated diagnostic platform for various organisms following the concept of public health laboratory given by Parson et al.3
Frontline Healthcare workers: In India, we have accredited social health activists (ASHA) working with community and the central government has recently started a course on frontline healthcare workers (akin to community health worker, CHW) as part of skill development. We propose to train these workers on animal health as well so as to make them One Health workers. They will also have access to smart phones with an innovative application for use of the CHW, medical officer and the local and central laboratories for maintaining health records of the patient using a biometric system, generation of requisition for various tests, tracking of samples shipped in a specially designed container to the reference laboratory, retrieval of results and prescription of treatment.

Newer Diagnostics for diseases of zoonotic importance: In tune of newer diagnostics for TB, we are also working on development of diagnostics tools for diseases common in our area, viz. malaria, brucellosis, scrub typhus and leptospirosis.
References:
Thierry Nyatanyi, Michael Wilkes, Haley McDermott, Serge Nzietchueng, Isidore Gafarasi, Antoine Mudakikwa, Jean Felix Kinani, Joseph Rukelibuga, Jared Omolo, Denise Mupfasoni, Adeline Kabeja, Jose Nyamusore, Julius Nziza, Jean Leonard Hakizimana, Julius Kamugisha, Richard Nkunda, Robert Kibuuka, Etienne Rugigana, Paul Farmer, Philip Cotton, AgnesBinagwaho. Implementing One Health as an integrated approach to health in Rwanda. BMJ Global Health Feb 2017, 2 (1) e000121; DOI: 10.1136/bmjgh-2016-000121
Narang R, Narang P, Deotale V. Containment Laboratory Running on Hybrid Power Sources: A Solution for Countries With Limited Access to Electricity? Int J Tuberc Lung Dis 2017; 21 (4), 480.
Parsons LM, Somoskövi Á, Gutierrez C, Lee E, Paramasivan CN, Abimiku AL, Spector S, Roscigno G, Nkengasong J. Laboratory diagnosis of tuberculosis in resource-poor countries: challenges and opportunities. Clinical microbiology reviews. 2011 Apr 1;24(2):314-50

Drs. Dadhich and Prasad raise important issues in their comments and questions about our study on home-based management of children with uncomplicated severe acute malnutrition (SAM).  We are grateful for the opportunity to respond to them.

In our view, there are two distinct goals in the management of children with SAM. The first, and more urgent, is to treat and move them away from severe malnutrition, a state of high risk of death. The second is to rehabilitate them to the point they are no longer malnourished. The treatments examined in the study achieved the first goal well. By the end of the treatment phase, 84.5% of the children in the RUTF-L group were no longer SAM. However, they did less well in achieving the second goal.  By the end of the treatment phase, 57% of those receiving RUTF-L had recovered to the point of no longer being malnourished. Should we have applied the same approach as most studies to calculate their WHZ - using height at enrolment and current weight, rather than concurrently measured height and weight - 81% of the children in the RUTF-L group would no longer be classified as malnourished by the end of the treatment phase.

We note Dr. Prasad’s misunderstanding that over 40% of the children in our study remained SAM by the end of the treatment with RUTF-L. Should her interpretation have been correct, it would justify her expressed concern. However, as stated above, by the end of the treatment phase 84.5% of the children in the RUTF-L group were no longer SAM. 

We also note the concern that the ease and extent of recovery in our study is less than reported in African trials. However, as indicated in the paper, the population in our trial was significantly different from that in African trials: only 0.2% of our children had oedema, while it ranged from 38-80% in African trials. The mean WHZ score in our trial population was -3.5 while the mean ranged from -2.0 to -2.8 in Africa.  Such differences make comparisons on ease and speed of recovery quite hazardous.  One informed expectation is that recovery in the Indian population would take longer as they were further from the target of WHZ ≥ -2 SD.

We acknowledge Dr. Dadhich’s interest in the comparison of the performance of RUTF-C and RUTF-L. However, the trial was not designed to compare them.  Given the lack of power for this comparison, we refrained from examining possible differences in their efficacy.  We know that the composition of the two products was essentially the same and believe that any potential differences would likely be due to users’ preferences.

We agree that additional ways need to be found to rehabilitate these children. We see that in responding to SAM needs actions at the hospital level for children with complicated SAM, and at the community level, for treatment of uncomplicated cases, rehabilitation and prevention.  Success requires that we provide treatment and also support families through comprehensive child care and promoting the best use of local resources.  The rehabilitation of malnourished children, the sustainability of improvements and prevention of SAM need greater attention and possibly new approaches. We may need to extend the period of treatment while scaling down the management with therapeutic foods, or we may be more successful by increasing efforts to improve complementary feeding while continuing to use therapeutic foods as a supplement, or we may focus our efforts on improving complementary feeding without recourse to supplements. These approaches – or other that may be identified – should be tested to guide future efforts.

In the end, rehabilitation by whatever means must be a part of a comprehensive program and reach all target infants and children. This will require engagement at home, community and facility and not only facility-based interaction. It may be that some redesign of our overall strategy is needed. The issue of nutrition and growth in early life is critical and needs an evidence-based, thoughtful and yet practical way forward.