Effect of implementation of Integrated Management of Neonatal and Childhood Illness programme on treatment seeking practices for morbidities in infants: cluster randomised trialBMJ 2014; 349 doi: https://doi.org/10.1136/bmj.g4988 (Published 29 August 2014) Cite this as: BMJ 2014;349:g4988
All rapid responses
Re: Effect of implementation of Integrated Management of Neonatal and Childhood Illness programme on treatment seeking practices for morbidities in infants: cluster randomised trial
We read the article titled “Effect of implementation of Integrated Management of Neonatal and Childhood Illness programme on treatment-seeking practices for morbidities in infants: cluster randomised trial”1.
Though there has been a steady decline in under-five mortality after IMNCI implementation in 2003, the rate of decline is not consistent with attainment of the Millennium Development Goal of reducing under-five mortality to 39 per 1000 by the year 20152. The reasons for shortfalls in achieving MDGs include deficiency in early identification of danger signs, delay in seeking care, and improper management of childhood illness. Strategic interventions that focus on reducing childhood mortality should address these critical points to achieve the desired goal. Hence, evaluation of a composite intervention strategy like IMNCI should also assess the impact on these key points. In this context, the current study has tried to find the effect of IMNCI programme on morbidity and treatment seeking behaviour. However, we would like to bring out few issues pertaining to the methodology of the study that may affect the interpretation of the results.
The authors intended to explore the effectiveness of routine IMNCI program on treatment seeking behaviour of the caregivers for neonatal and childhood illness. The impact evaluation studies try to attribute the causal relationship between the specific intervention and the desired outcome. Impact assessment requires comparison of the intervention group with a counterfactual which replicates the status of the outcomes in the absence of intervention3. The authors have chosen an ideal counterfactual by randomizing the catchment areas of Primary Health Centres within the same district into intervention and control arm. However, it is ideal to obtain baseline information on treatment seeking behaviour of caregivers, neonatal and childhood illness and hospital admission before the implementation of IMNCI intervention so that double difference estimates can be made.
The authors begin their discussion with a blanket statement “Following implementation of the Integrated Management of Neonatal and Childhood Illness programme, a substantial improvement was seen in treatment seeking and ...” By mere difference in outcomes between the intervention and the control clusters at end-line assessment, the authors have claimed a substantial improvement in outcomes. As impact evaluation studies try to attribute the observed difference to a specific intervention, difference-in-difference analysis would be a better statistical method, to identify the causal link between the implementation of IMNCI program and the pre-specified outcomes.
Apart from the routine incentive for promoting institutional delivery, the intervention has included additional incentives to community health workers for doing home visits, treating sick newborns and running women’s group meeting. These additional incentives are not part of the routine IMNCI program. Hence, the results shown in this study may be an over-estimate as expected from routine IMNCI program.
Additional monetary incentives do not create a lasting commitment among the community health workers4. Non-financial incentives like flexible working arrangements, positive working environments and access to benefits and support may prove better than financial incentives. Moreover, financial incentives demand long-term political commitment and financial resource allocation. So implementation of additional incentives into routine IMNCI programme may not be feasible in the current situation of health care in India.
1. Effect of implementation of Integrated Management of Neonatal and Childhood Illness programme on treatment seeking practices for morbidities in infants: cluster randomised trial | The BMJ [Internet]. [cited 2015 Jul 15]. Available from: http://www.bmj.com/content/349/bmj.g4988
2. Census of India : SRS Statistical Report 2013 [Internet]. [cited 2015 Jul 15]. Available from: http://www.censusindia.gov.in/vital_statistics/SRS_Reports_2013.html
3. Gertler PJ, Martinez S, Premand P, Rawlings LB, Vermeersch CMJ. Impact Evaluation in Practice [Internet]. The World Bank; 2010 [cited 2015 Jul 15]. Available from: http://elibrary.worldbank.org/doi/book/10.1596/978-0-8213-8541-8
4. WHO | Guidelines: incentives for health professionals [Internet]. WHO. [cited 2015 Jul 16]. Available from: http://www.who.int/workforcealliance/knowledge/resources/guidelines_heal...
Competing interests: No competing interests