This interest in costs and benefits has extended to Health Impact Assessment (HIA). Costs

must be explicit and proportional to the decision at hand. It would not be reasonable to

Environmental Impact Assessment Review 25 (2005) 791798

www.elsevier.com/locate/eiarDeveloping a framework to assess costs and benefits

of Health Impact Assessment

Phillip Atkinson a,*, Anthea Cooke b

aPublic Health Specialist, Southwark PCT, United KingdombIndependent HIA Consultant, United Kingdom

Abstract

This paper presents some early thinking as to how the costs and benefits of HIA might be

assessed. After considering previous work it uses a comprehensive HIA in Dulwich, SE London as a

case study to highlight the possibilities and difficulties of collecting necessary data on costs and

benefits. It then sets a context for developing a costbenefit framework for analysis. The framework

is viewed alongside the major types of economic evaluation. The paper concludes with a review of

outstanding issues and considers how evidence on cost and benefit might make a difference in the

application of HIA.

D 2005 Elsevier Inc. All rights reserved.

Keywords: Health Impact Assessment; Cost; Healthy Living Centre

1. Introduction

There is a growing national and local interest in the costs and benefits of public health

activities. A recent report on the future cost of healthcare in England (Wanless, 2004)

concluded dthere is generally little evidence about the cost-effectiveness of public health.TWith the need to meet targets and pressures on funding, local decision makers are

increasingly asking what difference an intervention will make and how much it will cost.0195-9255/$ -

doi:10.1016/j.

* Correspon

E-mail addding author.see front matter D 2005 Elsevier Inc. All rights reserved.

eiar.2005.07.011

ress: phil.atkinson@southwarkpct.nhs.uk (P. Atkinson).

spend o10,000 on an HIA to influence the use of o20,000 whereas it would be worthinvesting this sum to influence a decision that would affect many millions of pounds. The

since the HIA is attempting to estimate the impact, though effective scoping should

P. Atkinson, A. Cooke / Environmental Impact Assessment Review 25 (2005) 791798792provide a rough idea of the proposed effects. It has been argued that 10% of the cost of

health care interventions should be spent on evaluation and this figure might be applied to

HIA. The Finningley Airport HIA cost less than a tenth of the planning costs for this

1benefits of HIA are often implicit and rarely quantified. Evaluation has been identified as

the dweakest pointT of HIA and the need for information on costs and benefits of HIA isincreasingly recognised (Kemm and Parry, 2004b).

2. Cost and Benefits

The cost of undertaking an HIA should, in theory, be relatively easy to measure. Many

studies record attendees and the time devoted to meetings, community events and so on,

and it is usually possible to estimate the time taken to write the reports. Other costs, such

as external consultant time and catering, are known. With these figures together with the

cost of staff time it should be simple to calculate a total cost for the HIA. However in

practice the required information is often not available. Day rates and the time spent

(especially outside meetings) may be inadequately recorded and allocation of overheads

uncertain. External costs can be dlostT in other budgets (Kemm and Parry, 2004b). Forthese reasons the total cost of an HIA can be difficult to determine.

A small number of studies have attempted to evaluate HIA in the UK. These include

Alconbury Airport HIA (Close, 2001), Finningley Airport HIA (Abdel Aziz, 2003), the

review of four Mayoral strategies (London Health Commission, 2003). Other papers in

this edition add further examples (Elliott and Francis, 2005; Bekker et al., 2005). Very few

of these studies attempted to quantify the cost of each study and in many instances cost

was not even mentioned. The Alconbury evaluation (Close, 2001) estimated the time

devoted to the project, a total of 684 hours, but did not convert this into a monetary value,

and worked out that the dadditional expensesT (printing, conferences, expenses) wereo6000. The Finningley Airport evaluation (Abdel Aziz et al., 2004) took 348 person-days,equivalent to o52,20069,600 assuming an average cost of o150200 per day (includingoverheads). They spent o14,846 on outside consultants and around o2500 printing reportsso the total cost of the HIA was between o69,20086,600. Fleeman (1998) costed threeprojects within the Merseyside HIA programme some years ago at an average at o12,650each, and Ardern (2004) calculated the costs of a local transport plan HIA study as

o11,000, in addition to dcurrent work objectivesT.Little information is available to help practitioners estimate the cost of a proposed HIA

and until more published HIAs show their full costs it will remain difficult to quote costs for

those wishing to commission an HIA.1 It has been suggested that costs should be

proportionate to the size of the predicted impact. However this leads to a circular argumentThe London Health Observatory have recently developed a tool to estimate the cost of HIA. See

www.lho.org.uk.

project and a tiny fraction of the total cost of developing the new airport (Abdel Aziz et al.,

2004).

The benefits of an HIA are even more difficult to identify and to quantify. There are a

large number and wide range of benefits that need to be considered. Process benefits result

from undertaking the HIA and include wider appreciation of the determinants of health,

and enhancing individual skills. Impact benefits are evident very soon after completion of

the HIA. Of these, the most obvious and arguably the most important is the effect on

decision makers, but other impact benefits could include increased interagency working

and greater community involvement. Outcome benefits are delayed and are the

improvements in determinants of health or health itself that result from implementing

the recommendations.

Assessing and quantifying these benefits is challenging. The process benefits are the

easiest to assess whereas the impact benefits are more difficult. Assessing the effect of an

HIA on decision making is notoriously difficult. Questionnaires, interviews, focus groups

can all be used, but they rely on eliciting respondents views to (often narrow) questions and

shed little light on how people behave in practice (McDaid et al., 2002). Finally, assessment

of outcome benefits is more difficult still and so is often ignored in HIA evaluation. The

accuracy of the predicted effect on health is also rarely assessed (Parry and Stevens, 2001).

Different benefits will be evident at different stages. Process benefits become apparent

soon after the HIA, whereas impact benefits may only become assessable after months or

years and outcome benefits may only be realised after many years. Contamination is an

issue at each stage, as it becomes difficult to determine whether any benefits were due to

the HIA itself, or to other factors operating at the same time.

Those responsible for identifying and quantifying benefits will never be entirely

independent and objective, as all evaluation has a political component (Seedhouse,

1997). However, steps should be taken to reduce bias and ensure the use of measures

that are sensitive to each benefit. Assessing benefit can be a time consuming resource

intensive task and the use of this time and resource must be justified. Difficulties of

identifying and quantifying benefits are common to evaluation in general (Naidoo and

Wills, 2004).

A number of studies have attempted to assess the benefits of HIA (Jackson et al., 2005;

Elliot and Francis, 2005). Some interviewed stakeholders to identified perceived benefits.

Common themes included building knowledge and skills, among both participants and

authors; bringing clarity and presenting the whole picture as part of the decision making

process; putting health and public health onto the agenda of a wider range of agencies and

encouraging multi-disciplinary and interagency working. No study has attempted to

quantify these benefits.

Assessing cost and benefits is a complicated task requiring time and energy that are

often in short supply. Adding a substantial component to the already daunting task of

doing an HIA may be challenging. If health impact assessment is merged with other

impact assessments in integrated impact assessments it will become even more difficult to

distinguish costs and untangle benefits. Will evidence on cost and benefit of HIA make

any difference? It is difficult to tell. Unfortunately, evidence that information on costs and

P. Atkinson, A. Cooke / Environmental Impact Assessment Review 25 (2005) 791798 793benefits really influence decisions remains limited (Drummond, 1994; Sloan et al., 1997;

Hoffman and Graf von der Schulenburg, 2000).

3. The Healthy Living Centre in Dulwich

An HIA of a Healthy Living Centre will be used as an example to illustrate how

account can be taken of costs and benefits. A Healthy Living Centre (a centre providing an

integrated range of medical and other community services) was being proposed as part of

the plan to re-develop Dulwich Community Hospital, in South-East London. The multi-

agency steering group asked the local public health department to lead on a comprehensive

HIA to help development of the proposal. Scoping with the steering group supported by

P. Atkinson, A. Cooke / Environmental Impact Assessment Review 25 (2005) 791798794outside consultants started in July 2003, followed by a community profile, literature

review, stakeholder analysis focusing on older people, and substantial community

consultation. The final report was delivered in December that year. The recommendations

were accepted by the strategic planning group and are now being taken forward through

the business planning processes.

Initially it was assumed that assessing the cost of the HIA would present no real

problem, as detailed records were kept regarding expenditure and staff time devoted to the

different stages of the HIA but the task proved more difficult than expected. A decision

was taken to include the cost of all staff time, irrespective of the organisation they worked

for, to give a btotalQ cost rather than cost to the health service. This was the cost that shouldbe used to make comparisons with other public health interventions. However, for some

partner organisations staff salaries were not known and the individuals were not willing to

share this information. In these cases salary had to be estimated. Staff overheads were not

considered in these estimates, but in retrospect, should have been included. Total staff

costs were then calculated from the time that individual had devoted to the HIA and their

estimated hourly rate.

There were questions as to what contributions should be included. For example, should

the cost of the time of someone who simply attended one 2-h meeting be added to the

total? Should the costs of community participants be included? Over 200 people attended a

number of rapid HIA workshops, and were given no payment. Some might consider that

the value of their time should be included in the total cost of the project. This was not

done, but it should be considered in future HIAs. Besides staff other costs were largely

fees paid to external HIA consultants and a small amount to support meetings and publish

the various reports. These were relatively easy to quantify.

The costs for all the main components of the HIA, and the total cost, are shown in Table

1. The cost of meetings is largely made up of staff and consultant time but there is a small

element for room hire and catering. The rapid workshop with the steering group proved

Table 1

Main costs of the HIA of Dulwich Healthy Living Centre

Steering group and other meetings o2154Rapid workshop with steering group o2964Community profile o3800Literature review o4537Stakeholder consultation o2800Community consultation o5209Recommendations o3188

Total cost o25,000 approx

relatively expensive but this included preparation for all the other workshops used in the

community consultation later on. The community profile was compiled by a postgraduate

student under the supervision of a public health specialist, and the stakeholder consultation

by one of the external consultants. The literature review, identifying characteristics of

dsuccessfulT Healthy Living Centres and examining the effectiveness of proposed projectsin the centre itself was conducted partly by the external consultants and partly by a public

health staff member. The community consultation was the most expensive part of the

project. It involved 10 events held over 3 weeks attended by over 200 people, and required

substantial organisation, facilitation, and time to write up the results. Finally, the cost of

the recommendations includes agreeing them with the steering group, writing the final

report, and a small amount for publication of the report. The total cost of the HIA came to

approximately o25,000, including all staff costs both internal and external.A review of the potential benefits of the HIA produced six categories:

! Influence decision makers;! Improved interdisciplinary/interagency working;! Promoting health;! Enhance individual skills, knowledge and experience;! Encourage community involvement;! Informing and assisting future HIAs.

P. Atkinson, A. Cooke / Environmental Impact Assessment Review 25 (2005) 791798 795Table 2 summarises how these benefits could be measured and their cost. Influencing

decision makers is probably the most important benefit, but one of the most difficult to

Table 2

Categories of benefit, methods of measuring them and cost

Benefit category Measurement of benefit Cost

Influence decision makers Post-HIA questionnaire asking what difference

it made, shadow decision makers, monitor

further request from decision makers

Total cost of HIA

Interdisciplinary/interagency

working

Attendance at steering group meetings, look

through minutes, evidence of effective joint

working during and after HIA

Cost of timea of

people attending

meeting

Promoting health (better understanding

of health and its determinants, putting

public health on wider agenda)

Pre- and post-HIA questionnaire to members

of steering group exploring understanding

and practice

Part of cost of

timea devoted to

the HIA

Enhance individual skills,

knowledge and experience

Questionnaires and interviews to assess skills

before and after HIA, explore how skills were

used and what difference this made

Part of cost of

timea devoted to

the HIA

Encourage community involvement

(sense of value/inclusion, enhanced

control, social networks, involvement

of hard to reach groups)

Post-HIA workshop questionnaire, numbers

coming to later workshops (Cost must also

include time costa of workshop organisers,

room hire, refreshments, etc.

Cost of timea of

participants)

coming to

workshops

Wider benefit Numbers asking for information or

copies of HIA

Cost of printing

and distributing

reportsa Cost of time is sum of (hours spenthourly rate) for each person involved.

assess. Questionnaires or interviews have been suggested, but they will only provide a

partial picture, as described earlier. Pre- and post-HIA questionnaires might help to elicit

the difference that the HIA made. Observation of decision makers working (McDonald,

2002) is an approach that might be adopted for some future HIAs. It is too early to assess

the impact of the Dulwich HIA on decision making, but the early signs are promising.

Improved interdisciplinary and interagency working have been cited as important

benefits of HIA. These may simply happen during the HIA, or continue after the HIA

is completed. This could be assessed by studies focusing on what has changed as a

result of involvement in the HIA. Evidence of ongoing interdisciplinary working would

provide a more robust objective measure of the impact of the HIA. Organisations

represented at the steering group meetings included voluntary sector, council, private

finance initiative company, Primary Care trust (health authority) and local residents

brought together through the Dulwich HIA. These continued to meet together to lobby

for the proposed centre, to raise funds, and to work with the architects.

Promoting health through dimproving understanding of how health is affected bydifferent factors and by putting public health on the wider agendaT is another importantbenefit. This may occur in a professional or community setting. Housing professionals

may for example realise the importance of their work for health, and members of the

community may start to understand how their health is more than simply not feeling ill.

Measurement of this benefit will depend on the context. Evaluation questionnaires are

often used at the end of community consultation events and perhaps these could be

developed, to reveal if understanding has changed. The community consultation was

particularly beneficial, in improving general understanding of health, especially among

dhard to reachT groups. The Dulwich HIA involved meetings with a black elderly group,an Asian elderly group, a visually impaired group, and two youth groups with a large

number of ethnically diverse young people. Many participants claimed their under-

standing of health had improved.

Previous studies have highlighted how involvement in HIA can enhance skills,

knowledge and experience. A skills audit is suggested as part of the scoping stage, of

HIA but is not often conducted. Perhaps more emphasis could be placed on

measurement of skills before and after the HIA to determine change.

Encouraging community involvement comprises promoting feelings of inclusion,

enhancing control, building social networks and social capital, accessing hard to reach

groups and addressing inequality. Benefits in this group are difficult to quantify and

methods for the measurement of social capital have been hotly debated. Nevertheless, it

should prove possible to adopt some simple measures that focus on process. A post-

workshop questionnaire could ask about the sense of inclusion and the number of

workshop participants that turn up at the recommendations event could be used as an

indication of engagement in the HIA.

Another benefit of each HIA is guiding and informing future HIAs. The number of

requests for information and reports from those not directly involved in the Dulwich

HIA is one indication of this wider influence. The literature review from the Dulwich

HIA was requested by several authorities proposing to develop Healthy Living Centres

P. Atkinson, A. Cooke / Environmental Impact Assessment Review 25 (2005) 791798796in other parts of London. Following the HIA, the Primary Care Trust (Health

Authority) Board requested a formal cost-effectiveness analysis for investing in the

start to quantify the benefits in particular areas. Many have been calling for evidence on

cost, benefit, impact and effectiveness of HIA, but little has actually been done. Now is the

In the current UK political climate with its focus on performance, targets, and valuefor money, evidence of cost and benefit is increasingly required for all activities

including HIA. Cost and benefit data will be needed to satisfy health managers, but other

justification for the use of HIA will also be required. There will still be a need to lobby to

ensure that HIA moves from being a mere dplay-thingT (Kemm, 2000) and becomes atool that is routinely used in public health practice to analyse policy and inform

decisions.

References

Abdel Aziz MI, Radford J, McCabe J. The Finningley Airport HIA: a case study Chapter 25. In: Kemm J, Parry J,

Palmer S, editors. Health impact assessment. Oxford7 Oxford University Press; 2004.Ardern K. HIA: a practitioners view Chapter 10. In: Kemm J, Parry J, Palmer S, editors. Health impact

assessment. Oxford7 Oxford University Press; 2004.Bekker M, Putters K, van der Grinten. Evaluating the impact of HIA on urban reconstruction decision making;

Who manage whose risks? Environmental Impact Assessment Review 2005; 25: 75871 [this issue].

Close N. Alconbury airfield develop health impact assessment evaluation: report for Cambridgeshire Health

Authority. Anglia Clinical Audit and Effectiveness Team, Cambridge.

Drummond M. Evaluation of health technology: economic issues for health policy and policy issues for economic

appraisal. Soc Sci Med 1994;38:1593600.

Elliott E, Francis S. Making effective links to decision making: key challenges for health impact assessment.time to start exploring these issues, difficult though they may be, to build our knowledge

in these areas. Methods to assess the influence of HIA on decision making are urgently

needed, as are tools to assess the effect of participating in an HIA in the local community.

Thought must be given to assessing the long-term benefits, such as improvements in health

over longer periods. Little attention to date has been given to negative impacts, but these

must also not be forgotten despite the obvious temptation to focus on the positive.

4. ConclusionHealthy Living Centre, which the London School of Economics have now been

commissioned to do.

The costbenefit framework works at a relatively simple level, comparing costs for

groups of benefits. The health economists would call this a cost-consequence analysis

(CCA) that looks at a range of outcomes, each with its own dimension. The different

outcomes are subjectively weighed and compared with costs. The final result is therefore

less tidy and more complicated than when using a single measure of utility (such as

QALYs) is used, but arguably closer to the real world scenario (Hale et al., 2003).

This paper has discussed the difficulty of assessing the costs and benefits of HIA, but

shown that it is possible to make a start. Relatively simple measures of benefit, such as

skills audits or post-workshop questionnaires could be easily incorporated into HIAs, and

P. Atkinson, A. Cooke / Environmental Impact Assessment Review 25 (2005) 791798 797Environ Impact Asses Rev 2005;25:74757 [this issue].

Fleeman N. Estimated time and costs of health impact assessments. Liverpool Public Health Observatory; 1998.

Hale J, Cohen D, Ludbrook A, Phillips C, Duffy M, Perry-Langdon N. UK health promotion and health

economics forum Moving from evaluation into economic evaluation: a health economics manual for

programmes to improve health and well-being. Cardiff7 Health Promotion Wales; 2003. www.hpw.wales.gov.uk/English/topics/health%20economics%20manual/manual.pdf.

Hoffman C, Graf von der Schulenburg JM. The influence of economic evaluation studies on decision making. A

European survey. Health Policy 2000;52:17992.

Jackson N, Taylor L, Quigley N. Process of evaluation of HIAs in England (unpublished). London7 NHS HealthDevelopment Agency; 2005.

Kemm J. Building health impact assessment into decision making (rapid response). Br Med J 2000;24:1398

[May].

Kemm J, Parry J. Future directions for HIA Chapter 36. In: Kemm J, Parry J, Palmer S, editors. Health impact

assessment. Oxford7 Oxford University Press; 2004b.London Health Commission and Greater London Authority. Report on the qualitative evaluation of four health

impact assessments on draft mayoral strategies for London. London7 London Health Commission; 2003.McDaid D, Mossialos E, Mrazek MF. Making use of economic evaluation. Int J Risk Saf Med 2002;15:6774.

McDonald R. Using health economics in health services: rationing rationally? London7 Open University Press;2004.

Naidoo J, Wills J. Health promotion, foundation for practice. 2nd edition. London7 Balliere Tindall; 2004.Parry J, Stevens A. Prospective health impact assessment: pitfalls, problems, and possible ways forward. Br Med

P. Atkinson, A. Cooke / Environmental Impact Assessment Review 25 (2005) 791798798Further reading

Kemm J, Parry J. What is HIA? Chapter 1. In: Kemm J, Parry J, Palmer S, editors. Health impact assessment.

Oxford7 Oxford University Press; 2004a.J 2001;323:117782.

Seedhouse D. Health promotion: philosophy, prejudice and practice. London7 Wiley; 1997.Sloan FA, Whetten-Goldstein K, Wilson A. Hospital pharmacy decisions, cost containment, and the use of cost

effectiveness analysis. Soc Sci Med 1997;45:52333.

Wanless D. Securing good health for the whole population HM Treasury. www.hm-treasury.gov.uk/consultation_

and_legislation/consult_wanless04_final.cfm.

Developing a framework to assess costs and benefits of Health Impact AssessmentIntroductionCost and BenefitsThe Healthy Living Centre in DulwichConclusionReferences