Service Availability Mapping (SAM)

Description

Service availability mapping (SAM) has been developed to generate data on the availability of health services and the health system inputs and resources required for their success. SAM is a source of statistics on distribution of services and facility infrastructure, infection control procedures, services offered, staffing, laboratory facilities, essential medicines, commodities and supplies, and treatment guidelines. A key component of SAM is the association of each service delivery point with specific geo-coordinates, which enables mapping of facility distribution and comparison with geographic, population and epidemiological profiles. SAM does not directly assess quality of care, but it does generate information about the readiness of facilities to offer specific packages of health care, a key first step in ascertaining quality.

Key Elements

STRENGTHS OF THE TOOL The major strength of SAM is its high degree of standardization coupled with its relative simplicity of design, content, and implementation. SAM is designed for implementation by district health teams using hand-held data collection and geopositioning devices. Data collected are downloaded daily by the district teams and can be analysed on the spot, eliminating the need for separate data entry. Using the free HealthMapper software, district teams can generate local area maps of health facility, human resource and service availability and compare themselves to other districts. SAM can be implemented in the average district in a few days and does not call for extensive external technical assistance once the initial training has been completed and a supervisory plan has been put into place. Experience indicates a high degree of sustainability, with several countries investing their own resources for national roll-out following the initial, externally supported, data collection effort. SAM specifically addresses the need to generate data on both public and private health facilities.

LIMITATIONS OF THE TOOL There are two components to the approach, namely, a key informant interview and a census of facilities comprising visits to each facility (see below in the “Target and Instruments” section). The key informant survey based on interviews with district-level respondents has several drawbacks. Data quality may be poor if the district team is not well-informed. It is important to prepare respondents about the nature of the survey and the types of information needed, in order that district-level respondents can bring together the necessary documentation and prepare their responses before the interview. Failure to communicate in this way means that interviewees arrive unprepared and much useful information goes uncollected.

If the private sector is large, the government health team is unlikely to have sufficiently detailed documentation on private facility operations, unless there is a well-functioning facility registration system. Urban areas are a particular challenge because of the large number of private facilities: experience has shown that it is essential to identify and visit all facilities to make an accurate assessment of service availability.

The challenge in conducting the facility census component is to identify all health facilities. Obtaining access to private facilities for the interview requires careful planning and preparation and it is essential to establish contact with private sector associations and with governance or regulatory bodies where they exist. Where relevant questions on use of specific health facilities have been included in household surveys, these can provide a source to identify private facilities.

Special efforts are needed to include smaller private facilities, which are easy to miss especially in urban areas. In many developing countries, there is a high degree of informality and small clinics are not registered. These may include, dispensaries, maternity homes and some private vendors of pharmaceuticals that offer basic health services such as “antenatal care,” blood pressure and urine glucose measurements. In these circumstances, a decision needs to be taken in advance about which kinds of facilities should be included in the census. In practice, countries generally confine the facility census to more formal health care settings or those offering a minimum number of outpatient visits or in-patient beds.

Over time, as problems such as these are addressed, reporting becomes more comprehensive with subsequent rounds of censuses.

Facility censuses as currently implemented typically do not generate information on quality of care, community involvement, patient satisfaction, or details about staff such as educational background and pre-service training.

Principles and Steps for Using the Tool

PARTNERS National-level stakeholders should be involved from the beginning. These include sections of Ministries of Health responsible for monitoring and evaluation and programme delivery (including reproductive health and related programmes) development partners; national institutions; non-governmental organizations, user groups, and other interested parties. Partners’ involvement in adaptation and dissemination is important to reduce duplication and ensure buy-in.

TARGET SAM is made up of two interrelated components:

District key informant survey: The district questionnaire is used among district (or equivalent administrative unit) health teams and generates an initial assessment of service availability, human resources, medicines, commodities and supplies, and coverage of national programmes. It involves interviews with informed respondents – the district medical officer and the district health management team – complemented by reference to documentation available in the health district and at national level. This information is subsequently verified through visits to health facilities (see below, “Facility census”). Information is sought about all key programme areas according to national priorities.

WHO has developed a questionnaire, a data entry tool on a personal digital assistant and GPS software (Health Mapper). Interviews with health management teams in all districts provide data on the availability of health services and other health resources. Existing lists of health facilities are updated during fieldwork by generating geo-coordinates for all health facilities, including new facilities and the private sector (see below). The fieldwork focuses particularly on incorporating private sector facilities into the overall list, which is the basis for the health facility census.

Facility census: The facility census questionnaire is applied to all formal public and private health facilities in the community. It complements and validates information from interviews with district health officer management teams, generating more reliable data and verifying facilities’ readiness to deliver specific high-priority interventions. The facility census aims to include visits to all public and private health facilities in a country (or in the district, region or province studied). Such a census should become the basis for a national and subnational monitoring system of service delivery, supplemented by facility surveys and in-depth assessments to ascertain quality.

As a minimum a facility census includes information on the health infrastructure (public or private, facility type with GPS coordinates); availability and condition of physical infrastructure and communication capacity; health workforce; and the general status of drug supply, equipment and commodities. At little extra cost, the census can also generate data on the availability of key services such as MCH and IMCI, HIV, TB, malaria, and immunization. The questionnaire addresses the availability of specific services, availability of trained health workers and their exposure to recent training in particular interventions; the presence or otherwise of guidelines; availability of essential medicines, equipment, and diagnostic facilities. The precise detail depends on national priorities, but all countries are strongly encouraged to generate data on all major health priorities including maternal and child health, reproductive health, HIV/AIDS, tuberculosis, malaria, and infectious diseases. Generating a core set of data on a continuing basis ensures that the district can monitor service availability over time and compare service delivery from one district to another, and with other countries.

The key output of a facility census is a national database of health facilities, equipment and services offered. It provides an independent source of numbers and distribution of health workers which can be compared with other sources of information. Additional information can include, for example, the presence of health care workers on the day of visit. Comparisons between districts and regions provide valuable information about the distribution of services within the country. Information on minimum standards for key services can be used to provide information to programme planners.

METHODOLOGY The facility census comprises all facilities in a district. In large countries with many districts, it is not always practical to conduct a full facility census nationally. Instead it is more usual to conduct the facility census in a sample of districts (which may be grouped into clusters depending on degree of urbanization, income, remoteness, etc.). The facility census is rolled out to all other districts over a period of two or three years.

TIME REQUIRED A first step is a stakeholder workshop bringing together national and district level health managers and programme officers, to review and adapt the questionnaires for local relevance, and determine final content and approach. 
This preparatory phase can take around one month.

The time required for data collection will depend on the distances involved and the local proximity of data collection teams. In general, either regional or district-level teams country have been responsible for data collection. The advantage of this approach is that the facility census can become institutionalized as a regular component of district-level monitoring, increasing the chances of sustainability and use of the data collection for planning and management. As a general rule, data-collection teams comprise two individuals and a driver who complete the district-level questionnaires and then move on to applying the facility questionnaire in the selected districts. Under such circumstances, time for data collection can range from 2-8 weeks for the entire country. The period for data collection depends greatly on the size of the area to be covered, available transport, the quality of the communications network and weather conditions. If the data collection team can visit only one or two facilities a day, it probably makes sense to recruit additional 
teams, especially for the initial, baseline census which is likely to require longer 
than the follow-up visits.

The district questionnaire takes approximately 45 minutes to complete. The facility questionnaire can be completed in between 45 minutes and two hours, depending on the size of the facility. Data analysis and report writing generally take up to six weeks, full time.

HUMAN RESOURCES A coordinator, programme area experts, field supervisors and field teams are required for planning, adapting questionnaires and implementation. For data collection, two individuals and a driver are usually sufficient for the district-level questionnaire. Once data collection at the district level is finished, teams can be redistributed to districts where the facility questionnaire is being applied. Data collectors should have some experience in applying a survey, although novices can sometimes be teamed up with experienced colleagues; continuous oversight is needed, perhaps in the form of daily and weekly team meetings.

A second stakeholders’ workshop should review and comment on the interim report findings, which are often rather different from what routine health management information systems report, especially with regard to the presence of trained staff, the availability of medicines and supplies, and facilities’ ability to deliver care. Findings must be based on documented evidence to be convincing to all stakeholders. 
 

BUDGET LINE ITEMS

Components	Average	Max.	(Country)	Min.	(Country)
Stakeholder workshop	$7,436	$20,566	Nigeria	$1,678	Vietnam
Data collection training including pilot testing of questionnaires	$29,119	$120,112	Nigeria	$6,300	Swaziland
Data collection	$42,488	$118,577	Nigeria	$10,750	Swaziland
Health Mapper and Pendragon training	$8,473	$14,722	Ghana	$1,481	Tanzania
Report writing and dissemination	$14,877	$39,073	Tanzania	$3,785	Vietnam
Other (translation, editing, printing, internal consultants)	$15,851	$39,073	Swaziland	$5,000	Ghana
Supplies	$4,422	$4,422		$4,422
Communications	$500	$500		$500
External technical assistance (includes travel and per diem)	$33,000	$33,000		$33,000
Total	$156,166	$390,045		$66,916

Note: External technical assistance (TA): average three trips per country, average stay 15 working days (total). On-site TA was provided for questionnaire adaptation, final programming of the questionnaires into PDAs, and training. TA during fieldwork was provided at a distance and included periodic contact with country staff to monitor progress and addressing any technical issues with the PDAs and GPS units. National staff supervised the fieldwork.

Adaptation and Technical Assistance

EXTENT OF EXPERIENCE WITH THE TOOL Countries conducting SAMs include Nigeria, Tanzania (2006), Uganda (2004), Zambia (2005), Kenya (2005), Rwanda (2005), Botswana, Ethiopia and Ghana. Not all countries have implemented both components systematically across all districts.

RECOMMENDATIONS FOR ADAPTATION Adaptation by country stakeholders is a key component of SAM. However, WHO advises that countries try to maintain core elements of the standard questionnaire in order to be able to compare findings over time and with other countries. SAM is intended to meet the information needs of a number of programmes and to generate data on human resources, health infrastructure and service delivery needs of all key public health interventions. This maximizes the benefits of the tool for programme managers generally and makes most efficient use of limited resources. WHO does not recommend using SAM as a stand-alone tool to generate data about single programme areas as this creates risks of overlap and duplication in countries.

Samples of questionnaires used by this tool can be found at: 
the Bank of Sample Questionnaires

POSSIBILITIES FOR TECHNICAL ASSISTANCE AND CONTACT INFORMATIO
Contact Carla Abou-Zahr, World Health Organization, Geneva 
abouzahrc@who.int