The data for this study came from the trial conducted among pregnant women in Bijapur district in north Karnataka during December 2011 and March 2014 [14]. The trial was conducted by investigators from Gynuity Health Projects (New York, USA), University of Illinois at Chicago (USA), University of California, San Francisco (USA), KLE University’s Jawaharlal Nehru Medical College (Belgaum, India), and BLDE University’s Shri BM Patil Medical College (Bijapur, India).

The trial compared two community-level strategies for PPH management: primary prevention of PPH (administration of 600 mcg oral misoprostol to all women immediately after delivery) and secondary prevention of PPH (administration of 800 mcg sublingual misoprostol to women if blood loss within one hour after delivery reaches 350 ml). Historically, primary prevention was the only policy option that was being proposed and had been evaluated. Primary prevention is about preventing the onset of a health condition compared to secondary prevention which aims to detect the health condition early and then attempt to manage its impact. As such, primary prevention is preferable from an incidence point of view because it does not allow the health condition to occur thereby saving the individual from experiencing any consequences of ill health in the short or long term. However, the costs, and supply-chain burdens of primary prevention programmes are challenging [14]. The trial conducted in Karnataka was the first to propose a secondary prevention strategy of PPH and investigated the feasibility and effectiveness of the model of care in comparison to primary prevention.

The trial included deliveries conducted by Auxiliary Nurse Midwives (ANMs) at health sub-centres and women’s homes. The unit of randomization for the study was ANMs, and deliveries enrolled by each ANM constituted a cluster [14]. After considering the exclusion criteria e.g. high risk pregnancy, in active labour or refused to participate the study, there were 2,984 cases included in the cost analysis, of these 1,064 cases were in primary prevention group and 1,920 cases were in the secondary prevention group. Blood loss was measured for all women using a blood collection drape (Brasss-V Drapes, Excellent Fixable Drapes, Madurai, Tamil Nadu, India). In the primary prevention group, 1,061 (99.7 percent) women got misoprostol. In secondary prevention group, 92 (4.7 percent) women had postpartum bleeding more than 350 ml of which 90 (97.8 percent) received misoprostol.

Clinical outcomes did not differ significantly between the two arms in the trial; however, side effects were more common in the primary prevention group. Significantly more women in primary than secondary prevention experienced shivering after delivery (39.5 versus 9.0 percent, difference = -30.5, 95% CI -56.4 to -4.5) [14]. Only 4.2 and 2.2 percent of women reported moderate or severe shivering in primary and secondary groups, respectively (P = 0.151). Less than 1 percent of women in both groups described having ‘intolerable’ side effects [14].

Per person cost of misoprostol for each strategy was calculated using the ‘ingredient’ approach. All information related to training, drug use, side effects was taken from the trial data. The prices of drugs, supplies etc. were those faced by the government.

Apart from calculating the costs of the two strategies, we also calculated the cost of conducting normal deliveries at sub-centres. As there is dearth of information on the government spending on normal deliveries in sub-centres in India, we did a unit cost analysis of three randomly selected sub-centres in Bijapur district of Karnataka to find out the baseline cost–i.e. the amount government is already spending for conducting deliveries at sub-centres. We then added the additional cost for implementing the strategy–primary or secondary prevention of PPH.

The study was conducted from the government’s perspective. The Karnataka trial concludes that the secondary prevention strategy offers policymakers a feasible and practical approach to address PPH at the community level [14]. The objective of the costing study was to examine the cost difference (if any) in primary versus secondary prevention strategy of PPH using misoprostol. We chose to conduct the analysis from government perspective as along with the effectiveness of the strategy, the policymakers will also have the cost information for both strategies which will help them in decision making.

We undertook the cost analysis in two parts. The first deals with the estimation of cost of introducing misoprostol as primary or secondary prevention of PPH on a per patient basis using the trial data. In the second part, we extrapolated the cases and costs to a cohort of 14.9 million births (the number of births take place in sub-centres and home in India) [3] in each group and estimated the incremental cost to the government for both strategies.

Cost analysis of introducing misoprostol as primary or secondary prevention involved costs of training of ANMs and medical officers; cost of administering misoprostol (cost of the drug, drape and treatment of any side-effects) and cost involved in treatment of PPH.

In the second part of our analysis where we extrapolated the cases and costs to a cohort of 14.9 million births in each group, we used the proportionate number of a particular case in the sample to arrive at the number of possible cases in the population. For example, in the secondary prevention group out of 1,920 women, 90 women had blood loss of 350 ml or more after delivery. We used the fraction (90/1920 = 0.047) and multiplied by 14.9 million to get estimated number of cases in the hypothetical cohort (698,438 cases).

To calculate the incremental cost of introducing misoprostol as primary or secondary prevention of PPH, we first estimated the amount the government already spends for conducting deliveries at sub-centre. Standard costing methodology was followed to calculate the operating cost of sub-centres [15]. The average cost of normal delivery at the sub-centre was calculated using the ANM’s proportion of time spent for normal deliveries, the annual depreciated cost of capital items used for normal deliveries and the drugs and materials used for the same. ANM’s time spent for conducting deliveries and drugs and materials used for the same were collected by interviewing the ANMs during data collection from the sub-centres in July 2014. Capital cost included annual depreciated value of furniture and equipment used to conduct normal deliveries. A 3% discount rate was used [16] and country specific useful life-years for various capital items. Drugs and materials were the actuals used while conducting deliveries.

It should be noted in this context that the salaries, price of equipment used to calculate the costs of the two strategies and conducting normal deliveries were taken from the government health facilities. There could be regional variations, but we don’t expect huge variation. Further, the drugs and supplies prices used for the analysis were also taken from the approved price list of the government which is also generally standard across India.

The Institutional Review Board of the University of Illinois at Chicago, approved the protocol of the trial on 18 November 2010 and provided an updated approval on 5 September 2012. The Institutional Ethics Committee on Human Subjects Research at the Jawaharlal Nehru Medical College at KLE University, Belgaum, India, approved the protocol on 20 January 2011. The Health Ministry’s Screening Committee at the Indian Council of Medical Research, New Delhi, India, approved the protocol on 23 August 2011.

Average cost of conducting normal deliveries at sub-centres were estimated INR 29,205 (US$487) of which 59% was human resources cost, i.e. the time cost of the ANMs for conducting deliveries. 23% of total cost was for drugs and materials used while conducting normal deliveries and rest 17% was capital cost. The cost per delivery at sub-centre was estimated INR 308 (US$5).

Cost components of both the strategies using government perspective are presented in Table 1. In secondary prevention strategy as misoprostol was administered only to those having blood loss more than 350 ml, the cost of the drug is much lower. Overall, the average cost of the secondary prevention strategy is lower by INR 6 (US$ 0.1) as compared to primary prevention strategy using the government’s perspective.

We did the extrapolation for 14.9 million births to get enough numbers of different events for each arm. The cost per birth (either at home or at sub-centres in the cohort of 14.9 million births) was INR 308 (US$ 5.13) for both the groups. The average cost per birth after implementation of the strategy would be INR 430 (US$ 7.17) for the primary prevention group and INR 424 (US$ 7.07) for the secondary prevention group (Table 2). Therefore, the incremental cost per birth would be INR 122 (US$ 2.04) and INR 116 (US$ 1.93) for primary and secondary prevention groups respectively.