The Protocol for Health Care Needs Assessment

Authors of each health care needs assessment review were asked to adhere to a protocol to ensure consistency of reviews and, in some cases, to guide authors through the epidemiological approach to health care needs assessment.

The central idea behind the epidemiological model to health care needs assessment was that the 'need' for services for a population depends on that population's ability to benefit in terms of health gain from the service (including preventative through to rehabilitative services). The total ability to benefit of a population depends on how much of a problem there is (prevalence/ incidence and severity), and on what can be done about it (effectiveness). The relative priority assigned to different needs will also depend on costs.

There were two key audiences for the health care needs assessment review:

1. Health Authorities, who are responsible for drawing up Health Improvement Programmes, which are to be local strategies for health and health care 

2. Primary care groups, the most developed of which will be responsible for commissioning health services for their local population (typically, population of 100,000)as well as providing more integrated services themselves

The reports reflected primary, as well as secondary, care services.

The three key sections of the report concern epidemiology, existing services available including costs, and service effectiveness.

Advice given on how the review should be presented and the structure each review had to adhere to, is as follows:

Protocol Headings...

Executive Summary

This should be as user-friendly as possible for a wide health authority and clinical audience. It is intended that the executive summary should be able to stand on its own. It should use the headings 1 to 6 below.

1. Statement of the Problem/ introduction

Statement of the problem and its context mentioning major issues and controversies relevant to health care commissioners. The 'problem' would normally be (i) a set of symptoms or (ii) disease, but also occasionally (iii) an entire speciality, and exceptionally (iv) a treatment or procedure - for example major discrete surgical procedures.

The context of the disease will include the differential diagnosis of its presentation. This will allow the relevance for recognition and referral in primary care to be brought out.

Another contextual issue is the variety of different settings in which patients find themselves managed.

Use this section to give all ICD numbers, HRG and HBG codes, and OPCS operation codes involved.

2. Sub-categories

Set out sub-categories of the problem (disease) which are meaningful to purchasers and planners of health care. i.e. not too many, and recognising their different requirements for services. (For example, in the case of diabetes, distinguish IDDM, NIDDM, DM with minor complications, DM with major complications; or in the case of HIV, distinguish asymptomatic from symptomatic, drug abuser from non abuser; people with support from people without and so on). These sub-categories should inform subsequent sections of the report. They will normally be population/ patient based (as in the above examples), and may well be similar to Health Benefit Groups (HBGs).


  • Sub categories should recognise people who would benefit from prevention as well as treatment and rehabilitation. 
  • Where the interventions are dependent upon the severity of presentation, it is useful for severity to form part of the categorisation.
  • Sub categories should include very high resource use patients (if they are distinct) 
  • Sources of information should include both research literature and logic or professional consensus.

3. Prevalence and incidence

List the range of prevalence and incidence rates estimated for the problem as a whole and, as far as possible, for the sub-categories too. Where practicable, this should include the prevalence of 'indications' for health service intervention (i.e. what proportion of people with a disease would benefit from an intervention). Give details of how the incidence varies by age and sex, region, as well as by socio-economic and ethnic variables. Indicate the confidence limits, which should be placed around such estimates.


  • Epidemiological data should be presented using as denominators one of the following: 1,000, 10,000, 100,000 or 1 million. (The original design of the series describing typical districts of population of 250,000 is no longer appropriate with the multiple mergers that have taken place). 
  • Sources should include both epidemiological literature and official statistics.

4. Services available and their costs

Summarise and quantify the services currently available in terms of capacity, activity and costs. A range of current services should reflect both typical and extreme levels of service available in England, to prevent, diagnose, treat, care for, and rehabilitate the patient sub-categories in both primary and secondary care (and elsewhere e.g. local authorities). A flow chart should be drawn to summarise the services likely to be used by the various sub-groups of patients. Quantification of services available should set out averages and ranges (national, district and practice levels).

Services available should cover structure (typical staffing, bed and cost levels) as well as processes and flows of patients. The section can also be used to note norms (whether national or regional). Variations in service utilisation and/or service availability should be included.

Available data on the unit costs of each service should be included.


  • The purpose of this section is to describe existing services, not to set out desirable services. 
  • Unit costs (approximate) of services are required. Cost and price data should be labelled with the year to which the data are applicable to allow readers to make allowances for inflation. 
  • The National Casemix Office (now  the NHS Information Authority) has agreed to provide routine data relating to Health Benefit and Health Care Resource Groups. 
  • 'Invisible' services that contribute to the whole (e.g. pathology services) should be noted. 
  • It may be helpful to sub-divide services into primary and secondary. Given the fluid nature of the boundary, these sub-divisions do not need to be rigid. Nevertheless, it is worth distinguishing between specialist and non-specialist services which may be a way of distinguishing what can and can't be done in primary care (a question of numbers and economies of scale as well as skills). 
  • Sources should include literature, informal surveys, and official statistics.

5. Effectiveness of services and interventions

Summarise what is known about the effectiveness of the main services (outlined in section 4 above) and interventions involved. Where nothing is known about the effectiveness of the treatment or care, please indicate. Similarly, indicate any treatments that are known to be ineffective. This section should include any anticipated changes in the effectiveness of services, such as technological developments. Effectiveness of expensive new drugs and technologies should be included. Estimates of the effectiveness should be graded as in the scale shown in the appendix, i.e., size of effect and quality of supporting evidence. This section should give some indication of appropriateness, including who should undertake interventions (distinguishing between primary and secondary health care as appropriate).

The main source will be the literature on effectiveness, especially systematic reviews and also costs (described in section 4) to generate cost-effectiveness. As a rule, the literature on cost-effectiveness as such is very limited, but should be included where available. Cost effectiveness should be approached by melding of separate effectiveness and cost data.

6. Quantified Models of Care/ Recommendations

In this section authors should build on the above sections in order to outline what they think the service should look like. It should include quantified recommendations to meet needs for the specified patient sub-categories and the number of patients likely to be involved. Ideally all recommendations should be justified on cost effectiveness grounds. At the very least please distinguish between recommendations that are cost neutral and cost increasing. It is acceptable to put forward alternative models of care, which may have different implications for the primary/secondary interface. Models that require increased resources should indicate the costs involved.

This will in effect be your version of what a National Service Framework and a Local Health Improvement Programme for the topic area might look like. It is envisaged that the National Service Frameworks will set out what pattern and level of service should be provided for patients with a given condition.

7. Outcome measures and audit methods, targets and, 8. Information and research requirements.

These are all intended to be relatively minor headings, and are not essential to the main body of the text.

Consider what outcome measures and audit methods might practically be used to monitor services. Ideally proposed measures should be capable of assessing the benefits of treatment/ care to the patient but include process indicators that are relevant to understanding whether services are adequate.

Consider the relevance of any existing or likely targets for care in relation to this topic, and how such targets might be made most useful. Targets will be set by Health Authorities for the Primary Care Groups, and built into service agreements that Primary Care Groups make with NHS Trusts.

  • Indicate any changes in information required to:
  • facilitate needs assessment studies at local level 
  • enable future targets to be met 
  • enable outcomes to be monitored.

Indicate any research requirements arising from the review of needs assessment. Indicate where research investment is likely to lead to significant payback in terms of informing what would be optimal patterns of service provision.

General points

Objectivity  The review should be objective in the description of existing services and effectiveness. Desirable standards and norms should be restricted to the recommendations section.

Length  Please be as concise as possible. No more than 50 sides in the main report. Detailed information may be given in Appendices.

Quantification  Particularly in describing existing services (section 4) and recommended changes (section 6) please quantify services e.g. approximate staff numbers, beds, cases etc.

Formatting  Referencing should be in the Vancouver style - as per BMJ references.

Publication  As it is intended that your document will be published together with other in the series, please avoid material explaining the purpose of the whole series and the meaning of needs assessment etc. This is dealt with in the introduction to the series as a whole.

Please send the final version on disk, with all tables and figures/ diagrams filed in separate documents, and not within the text.



A. The procedure/service has a strong beneficial effect.

B. The procedure/service has a moderate beneficial effect.

C. The procedure/service has a measurable beneficial effect.

D. The procedure/service has no measurable beneficial effect.

E. The harms of the procedure/service outweigh its benefits.


I-1 Evidence from several consistent, or one large, randomised controlled trial.

I-2 Evidence obtained from at least one properly designed randomised controlled trial.

II-1 Evidence obtained from well-designed controlled trials without randomisation, or from well designed cohort or case-control analytic studies.

II-2 Evidence obtained from multiple time series with or without the intervention. Dramatic results in uncontrolled experiments (such as the results of the introduction of Penicillin treatment in the 1940s) could also be regarded as this type of evidence.

III  Opinions of respected authorities, based on clinical experience, descriptive studies, or reports of expert committees.

IV Evidence inadequate and conflicting.