Estimating the needs of populations

Service needs of populations depend to a large extent upon the frequency with which conditions occur in that particular population. Epidemiological methods provide estimates of the incidence and prevalence of disorders in communities and also data that can be used as measures of outcome, to assess whether or not the population's needs are being adequately met (see ChapterZ.^).

Community surveys provide the most complete information on disorders in the community and the characteristics of people who have specific illnesses. A notable example is the Epidemiologic Catchment Area study in the United States, which sampled a total of 18 571 individuals drawn from randomly selected households in four major cities and one mixed urban-rural area.(23) A standard diagnostic tool, the Diagnostic Interview Schedule, was used to collect data from these subjects, which permitted prevalence estimates to be made for 13 psychiatric conditions. A more recent American data set is provided by the National Comorbidity Study, which sampled people across the entire United States.(24) This study employed an internationally developed instrument, the Composite International Diagnostic Interview. (25> The British Psychiatric Morbidity Study(26> used a nationwide sample in Great Britain to make estimates of the frequency of psychiatric disorders diagnosed with the Schedules for Clinical Assessment in Neuropsychiatry.(27)

Prevalence estimates derived from surveys such as these are not in themselves sufficient indicators of need, but must be interpreted in the light of other data. As has been noted above, symptom severity, functional impairment, the availability of family and other supports, comorbidity, and dangerousness all affect a person's level of need. True needs estimates, therefore, must provide additional information about other factors that will influence the array of services needed. Lin et al.(28 make this point and have developed a summary need index including diagnosis, disability, and vulnerability factors to predict need for treatment, rehabilitative or preventive services. They also emphasize that need for mental health services is not to be considered as an all-or-none phenomenon, but as a graded variable; some people, even some of those with severe mental illnesses have few needs, others have many needs.

Using methods such as these, epidemiological studies frequently demonstrate that as many as half of those people who have major psychiatric disorders and other indicators of probable need are not in fact receiving mental health services. (29) This provides a measure of unmet need.

Case registers are tools used by epidemiologists for recording data on every person receiving services within a specific area. (30 A case-register data set can be quite simple, or it can include an extensive set of demographic and diagnostic information. From the perspective of needs assessment, a principal drawback is that case registers only provide information about those people who received services, and generally only information about those people who receive services within the specialty sector. They do not provide information about those who, for whatever reason, did not receive services, or received them within the primary care sector. Another limitation of case registers is that information is submitted by clinicians and does not usually provide 'research quality' data. Diagnoses, for example, may be idiosyncratic or inaccurate. Case registers require that individual patients are clearly identified, so that duplication will not occur and episodes of care that are provided in different facilities are attributed to the correct person. This requires the service provider to give detailed information to an external agency and thus constitutes a threat to privacy and confidentiality. This, along with the high cost of maintaining a register, has severely curtailed the application of this method. In the past, case registers have provided very useful information on treated patients, but almost all have now ceased operation.

Management information systems and government databases, maintained by health departments or ministries, regulatory bodies, third-party payers, and other agencies, provide data on utilization. These databases contain information about diagnosis, treatment contacts, and demographic characteristics, without identifying individual patients.

Such data may be considered to reflect demand, as discussed above, but they provide only an indirect measure of need. However, comparisons can be made between geographical areas or population groups to provide suggestive evidence of differences in practice patterns or unmet need. Data such as these are sometimes used to provide norms for planning purposes in the absence of true indicators of need: for example, in deciding how many hospital beds are needed for an urban or rural community.

Prevalence estimates for particular areas can be made using data from existing surveys, adjusted to reflect the demographic characteristics of the community of concern. Combined with demand projections, this can provide a basis for projecting likely service needs for the community, be it a local catchment area, a county, or a state. This method is employed in the United States to provide prevalence estimates of serious mental illness for state community mental health services planning, as is required by law. Serious mental illness is defined in terms of illness and disability, (31) which provides a better indicator of need than diagnosis alone. Data on diagnoses and functional impairment from the ECA study and the NCS were used to make estimates of serious mental illness morbidity by the following method: a prevalence estimate was made for each census tract in the United States after adjusting for demographic and economic variables. (For enumeration purposes the United States is divided into 61 253 census tracts; demographic data from the decenniel censuses are available for each of these tracts.) These estimates were then aggregated to provide state level estimates and national estimates, to be used for planning or administrative purposes. The total prevalence of serious mental illness in the country as a whole was estimated by this method at 10 million (5.4 per cent). (39

Booze Basher

Booze Basher

Get All The Support And Guidance You Need To Permanently STOP The Battle With Alcohol Once And For All. This Book Is One Of The Most Valuable Resources In The World When It Comes To Transformational Tools For Battling Booze Binges And Staying Alcohol-Free.

Get My Free Ebook


Post a comment