# Cost Effective Analysis (CEA)

• Usually, once a particular welfare policy is selected after CBA, Cost effective analysis is used to identify the least expensive method of attaining that definite result.
– CEA involves choosing one among different possible ways for achieving the desired result.
– E.g. it was calculated that ‘reduction in indoor air pollution’ would result in greatest economic benefit (after CBA) to the population, various possible ways of achieving this are analyzed using cost – effective analysis
– The one found to be the most efficient (highest cost effective) may be decided upon.
• CEA may be used even at still lower levels for individual intervention too, e.g. comparison of the costs of two or more different models of bio–gas plant
The output is measured in terms of health units and NOT money
Examples:
– No. of household with reduced IAP per 1000 ₹ spent
– No. of children immunized per 1000 ₹ spent
– Commonly used outcome is ‘no. of DALYs’ gained per unit cost reflecting a weighted combination of mortality and morbidity effects of an intervention
– Other possible outcomes are ‘cost per life saved’ or ‘no. of lives saved per unit cost’ for different treatments
• Sometimes, a population model is required to be constructed for accurate estimation of the effectiveness
Example of CEA: Comparing the Cost Effectiveness of Two Approaches for Immunization Clinic
• Approach A: the health worker makes house to house (H to H) reminder visits to bring child for immunization. Tally how many of them would attend the immunization session– let’s say it’s 43 kids are estimated
• Approach B: send reminder SMS to attend immunization clinic. Estimate how many are likely to attend. Let’s say that it is estimated at 31 kids.
• The next step is to calculate the cost of each activity.
– Take care that every cost associated with the activity is included, e.g. the cost of the time, transport etc.
– Let’s say that the total cost of house to house visits was ₹ 3,500
– The total cost of the SMS messages was ₹ 700.
• Now divide the total cost of H to H visits and of SMS messages by the total number of mothers who were calculated to come to the immunization clinic.
– H to H visits strategy = 3500/43 = ₹ 83.4per child
– Mass SMS: 700/31= ₹ 22.6 per child
• In this hypothetical example, the SMS messages are more cost effective that one-on-one outreach when it comes to mothers bringing the child to the immunization clinic.
• Often one needs to construct a population model instead of actually carrying out the study e.g. How many households use cellphones , cost of service providers, literacy etc.
CEA- Another Hypothetical Example: Comparison among Types of Clean Fuels
• Following are some fuels available for replacing the polluting coal/wood burning (for reduction of indoor air pollution)
– Kerosene
– LPG cylinders
– PNG
– Electric stoves
• The population model will consider the household composition (number, age and sex groups), type of houses, SES etc.
• Studies show that
– LPG use can save 75,630, 000 DALYs if 95% of population is covered
– Kerosene use also saves an equal no. of DALYs
– An improved stove which directs the pollutants out saves 51,540,000 DALYs
– Bio gas use can add 76,670, 000 DALYs
– Use of electricity saves the maximum no. of DALYs i.e 99, 200, 700
– The cost of bio gas plant was similar in both urban and rural areas but
– In urban areas only a few pockets exist which have an environment conducive to building bio-gas plants, one plant will benefit, let’s say 140 households with an average of 5 members
– In rural areas, one plant can provide clean fuel to almost 95% households with average 7 members each.
• In rural areas it was calculated that:
– For each ₹ 1000 investment in LPG and gas pipeline, the DALY saved were 51,230 per district
– For each ₹ 1000 investment in bio gas plant the DALY saved were 75,670 per district
– Supplying electricity to rural homes requires extensive infrastructure, and the cost of its use as fuel is also prohibitive, and the for each ₹ 1000 spent, only 100 DALYs were added despite it being the cleanest fuel.
• In urban areas it was calculated that:
– For each ₹ 1000 investment in LPG and gas pipeline, the DALY saved were 91,290 per ward
– For each ₹ 1000 investment in bio gas plant the DALY saved were 15,670 per ward
– For each ₹ 1000 investment in electric stoves, the DALY saved were just 5,430 per ward
– The cost of covering urban population with LPG and gas pipeline was much lower than covering a similar population in villages (rural)
• Hence bio gas plants are more cost effective in rural areas and LPG and gas pipeline are more cost effective in urban areas
Uses of CEA
• For selecting the most efficient intervention for achieving the objectives.
– As CEA estimates the costs and health gains of alternative interventions
• For evaluation of various activities under a program:
– During evaluation, the cost-effectiveness of the existing intervention is compared with the cost-effectiveness of a new proposed intervention or with others reported in the literature
– A decision to continue or replace the intervention can then be deliberated
– CEA is easier to calculate than cost-benefit ratio due to the fact that less information is required and that this information is more accessible
Limitation of CEA
– CEA doesn’t make comparisons between the interventions producing different outcomes
• Hence can only be used for ranking interventions with a common outcome, not for decision since it does not state if the benefits of the measure exceeds the costs; like CBA does
References:
• WHO, 1974. Modern Management Methods and the Organization of Health Services; Geneva
• WHO, 2006 . Guidelines for conducting cost–benefit analysis of household energy and health interventions; eds: Guy Hutton,Eva Rehfuess. WHO press, Geneva.
• WHO, 2003. Who Guide to Cost- Effectiveness Analysis; Geneva
• Park’s Textbook of Preventive and Community Medicine. 24th ed, 2017, Bhanot Publishers, Jabalpur
• MIKAEL S, LARS H; A Comparison of Cost-Benefit and Cost-Effectiveness Analysis in Practice: Divergent Policy Practices in Sweden; Nordic Journal of Health Economics, Vol. 5 (2017), No. 2, pp. 41-53