Setting Priorities Among Diseases and Conditions

After identifying a particular reference population in need of intervention, organizers of community screening programs must decide what health conditions or diseases should be given priority for new or expanded efforts. Potential criteria for ranking conditions or diseases in order of priority include disease burden; availability of one or more effective screening tests; availability of effective treatment and follow-up for cases detected early; cost-effectiveness, initial costs, and availability of resources for treatment and follow-up; community consensus about preferences, urgency, and equity; and historical precedent (i.e., experiences with screening programs already in place) (USDHHS 1991; USPSTF 1996).

The relative weight given to each criterion varies among communities and decision-makers and often must be negotiated among the involved parties. Some practitioners believe that the relationship between the rarity (prevalence) of a candidate disease and the proportion of false positives (persons without the target condition who test positive, given fixed sensitivity and specificity) is the most important principle of screening. This principle explains why selective screening strategies and risk-based strategies are often preferable to mass screening. Moreover, sequential screening to reduce false positives in the case of a rare disease increases in importance when the adverse consequences of a false positive test are severe and costly. In the final analysis, therefore, conditions of low morbidity and mortality (e.g., lower back pain) tend to be unsuitable candidates for screening even when highly prevalent. Similarly, conditions of high morbidity and mortality (e.g., HIV ) tend to be better candidates for screening efforts, even when they are rare (low prevalence).

Important epidemiologic concepts and principles that relate to setting priorities among health conditions eligible for screening are natural history and clinical stages of disease and burden of suffering (e.g., incidence, prevalence, mortality, and case-fatality rates; quality of life; and cost of illness). Often, a public health agency must decide whether expansion of the activities of an existing screening program is preferrable to starting a new screening program for another condition. In this instance, the same concepts and principles are germane to evaluating the effectiveness of the screening program that has already been implemented; past performance becomes relevant to the issue of expansion.

Natural History of a Disease. The natural history of a disease in a population is a sequential record over time of pathogenesis from initiation by one or more causal agents through clinical manifestations as signs and symptoms, the usual circumstances of its presentation and diagnosis in routine medical care, the rate of progression and response to treatment, and ultimate health outcomes. The term "natural history" also is sometimes used in the practice of epidemiologic surveillance to refer to trends in morbidity from a disease over time (Last 1995).

Lead Time. Figure 7-2 illustrates the natural history of a hypothetical case of hypertension, which is a target of screening both as a preventable disease and a risk factor for cardiovascular disease. Biologic onset is the point at which interactions between human host gene products, exposure to causal environmental agents, and behavioral risk factors lead to the initiation of a pathologic process (e.g., carcinogenesis, atherosclerosis, or physiologic dysfunction). The earliest point of detection of preclinical or asymptomatic disease is the point in disease progression at which an available screening test becomes effective. The usual point of diagnosis, in the absence of a screening program, is the point at which affected persons seek medical care for symptoms or signs

Time

Treatment Outcome

-cure f

Preclinical complications

- heart

- cerebrovascular

- renal

- remission

- disability

- death

Normal

Lead time-

Biologic onset

- normal blood pressure

Usual point of clinical diagnosis without screening

- high blood pressure

- heart disease

- stroke

- kidney disease

Preclinical disease

- high blood pressure

Figure 7-2. Natural history of a hypothetical case of hypertension of complications (e.g., pain, bleeding, or a mass) or a motivated clinician initiates a diagnostic procedure for some other reason (Morrison 1992; Sack-ett et al. 1991). Lead time is the interval from the earliest point of detection by a screening test to the usual point of diagnosis in the absence of screening.

The natural history of a disease determines its suitability for screening. In addition, effective screening and treatment alter the natural history of a disease in both individual cases and in populations by improving health outcomes. The opportunity to control a disease by means of screening and treatment will arise only if the disease progresses through an asymptomatic phase during which it is usually undiagnosed but detectable. Moreover, early treatment must prolong life or reduce morbidity more effectively than later treatment to make intervention worthwhile (Morrison 1992). In the absence of a screening program, the point of usual clinical diagnosis is determined by the interaction of levels of popular awareness of symptoms and signs, provider awareness and incentives to act, care-seeking behavior of the population, and access to medical care (Morrison 1992).

Clinical Stage (or Severity of Illness). The clinical course of a disease is that part of its natural history that begins at diagnosis and ends at recovery, death, or disability (Sackett et al. 1991). The short-term objective of screening is to shift the distribution of cases detected by screening toward less advanced clinical stages or severity of illness at diagnosis when prognosis after treatment is better. For example, the short-term goal of breast-cancer screening in a community is to increase the proportion of cases detected by screening that are localized or limited to regional spread (CDC 1996b; Kosary et al. 1996). Similarly, the short-term goal of screening for high blood pressure is to detect persons with hypertension, particularly those with more severe disease, and to reduce blood pressure to safer levels. The clinical stages of breast cancer and hypertension are illustrated in Tables 7-1 and 7-2, respectively.

Burden of Suffering. The burden of suffering in a population from a health condition or disease is an indication of its public health importance. Measures of disease burden include incidence rate, prevalence rate, mortality rate, case-fatality rate, quality of life, and cost of illness. High incidence rate (IR) (defined in Chapter 2) usually indicates a need for screening services. Effective screening and treatment for a disease, however, also alter its incidence rate in the screened population (Morrison 1992). Thus, trends in the IR of the disease may be useful in evaluating the uptake and effectiveness of a screening program. During periods of increasing use of a new screening test for a disease, the incidence rate of disease may increase, decrease, level off as uptake of the test becomes optimal, and finally decrease as preclinical cases are exhausted. For example, the age-adjusted breast-cancer incidence rate among women (of

Table 7-1. Clinical Staging of Breast Cancer by the TNM and SEER Systems and Mammography Coding by the Karolinska System

Stage Grouping (TNM System)a SEER Karolinska Mammography Coding

Systemb System (.Nonpalpable Cancers)c

Stage 0 (carcinoma in situ)

Unstaged

Stage I (T^2 cm; no nodes; no distant metastasis)

Stage IIA (T<5 cm; ± axillary nodes; no distant metastasis)

Stage IIB (T2-5 + cm; axillary nodes; no distant metastasis)

Stage IIIA (T any size; + axillary nodes; no distant metastasis)

Stage IIIB (T any size + extension to chest wall or skin; + internal mammary nodes; no distant metastasis)

Stage IV (T any size; nodes; distant metastasis)

Localized Regional

Regional Regional

Regional Distant

1 = normal mammogram

2 = abnormal—cancer not suspected

3 = abnormal—cancer suspected

4 = probable cancer

5 = definite cancer

'Source: Kinne (1991): T = primary tumor size at greatest diameter; N = node; M = metastasis. bSomce: Kosary et al. (1996): Surveillance, Epidemiology, and End Result. Localized is an invasive neoplasm confined to the organ of origin. Regional is a neoplasm that has extended beyond the organ of origin directly into the surrounding organs, tissues, and/or regional lymph nodes. Distant is a neoplasm that has spread to parts of the body remote from the primary tumor either by direct extension or by discontinuous metastasis.

cSource: Svane et al. (1993): Mammograms classified into codes 3-5 (<3%) are recalled for follow-up studies including fine needle aspiration, cytology, and/or biopsy.

all races) increased 32 % during 1980-1987, partly resulting from increases in early diagnosis and use of mammography as a screening examination (Miller 1993). During 1987-1992, breast-cancer, incidence remained relatively stable (NCHS 1996a).

A high prevalence rate (defined in Chapter 2) also is a measure of need for screening. In the absence of widespread screening for a disease, the population prevalence rate reflects the detection of symptomatic or clinical cases and the effectiveness of treatment in preventing death and achieving cure. In the presence of screening, the prevalence rate also depends on the periodicity of repeat screening, the mix of clinical and preclinical cases, and rates of death and cure. Because screening aims to detect preclinical disease, the prevalence rate of diagnosed preclinical disease increases as the uptake of screening increases in a population (Morrison 1992). The population prevalence of the target condition is a factor that determines the predictive value positive (PVP) of a screening test (see section on predictive value of a positive test). If sensitivity and specificity are constant, the higher the prevalence rate, the higher the PVP (Morrison 1992).

High mortality rate from a potentially lethal condition is perhaps the best indicator of need for a screening program. The population mortality rate or death rate from a disease is analogous to the incidence rate, except that the event being counted in the numerator is death instead of onset of disease (Kelsey et al. 1996). Death rate is one of the most useful measures of the value of a screening program that aims to prevent or delay death by early detection and intervention to favorably alter the natural history of the disease (Morrison 1992). Screening for hypertension aims to reduce the population mortality from stroke and heart disease (NHBPEP 1993), and screening for breast cancer aims to reduce the population mortality rate from that condition (CDC 1996a). The case-fatality rate (CFR) is the mortality or death rate among persons diagnosed with the disease. This rate is the complement (CFR = 1 - SR) of the survival ratio (SR)—the proportion of diagnosed cases who are alive at the end of a defined time period since diagnosis (usually 5 years) (Morrison 1992). The CFR and SR overestimate the effect of screening and treatment in screened populations (compared with unscreened populations) because of lead time and prognostic selection or length biases (see section on choosing effective screening tests). Therefore, they are

Table 7-2. Clinical Staging of Hypertension for Adults Age 18 Years and Older3

Category

Systolic (mmHg)

Diastolic (mmHg)

Normalb

< 130

< 85

High normal

130-139

85-89

Hypertensionc

Stage 1 (mild)

140-159

90-99

Stage 2 (moderate)

160-179

100-109

Stage 3 (severe)

180-209

110-119

Stage 4 (very severe)

a210

is 120

«Not taking antihypertensive drugs and not acutely ill. When systolic and diastolic pressures fall into different categories, the higher category should be selected to classify the individual's blood pressure status.

'Optimal blood pressure with respect to cardiovascular risk is SBP <120 mmHg and DBP <80 mmHg. However, unusually low readings should be evaluated for clinical significance.

'Based on average of two or more readings taken at each of two or more visits following an initial screening. Note: In addition to classifying stages of hypertension based on average blood pressure levels, the clinician should specify presence or absence of target-organ disease and additional risk factors. Source: National High Blood Pressure Education Program (1993).

«Not taking antihypertensive drugs and not acutely ill. When systolic and diastolic pressures fall into different categories, the higher category should be selected to classify the individual's blood pressure status.

'Optimal blood pressure with respect to cardiovascular risk is SBP <120 mmHg and DBP <80 mmHg. However, unusually low readings should be evaluated for clinical significance.

'Based on average of two or more readings taken at each of two or more visits following an initial screening. Note: In addition to classifying stages of hypertension based on average blood pressure levels, the clinician should specify presence or absence of target-organ disease and additional risk factors. Source: National High Blood Pressure Education Program (1993).

unsuitable outcome measures of the efficacy or effectiveness of screening programs.

The primary long-term goal of screening is to prevent or postpone death from a particular target condition. Life recouped by postponing death varies in both quantity (years) and quality (absence of impairment or disability). Quality of life measures, population health-status measures that combine the quality or desirability of a health state with its duration, are discussed in Chapter 8. A variety of survey instruments have been developed to collect the data needed to construct quality of life indexes (e.g., Quality of Well-Being [QWB] Scale) (Dasbach and Teutsch 1996). The number of quality-adjusted years of life saved by a screening program often is used as the effect measure when comparing screening programs that affect both morbidity and mortality, aim to improve social or physiologic functioning as a primary objective, or lead to outcomes measured in dissimilar natural units of morbidity or mortality. For example, a quality of life measure would be a more suitable outcome measure for comparing the relative cost-effectiveness of two screening programs for neural tube defects (to prevent mental retardation and death) and perinatal HIV infection (to prevent opportunistic cancers and infection, and death), respectively (Dasbach and Teutsch 1996).

In addition to measures of morbidity and mortality, the population burden of illness can be assessed in terms of the economic cost of a target disease for which screening is being considered. Health conditions with higher cost-burdens are usually given higher priority for developing and implementing screening technology. Approaches to assessing the economic cost of illness are discussed in Chapter 9 (Haddix et al. 1996; Luce et al. 1996).

Reducing Blood Pressure Naturally

Reducing Blood Pressure Naturally

Do You Suffer From High Blood Pressure? Do You Feel Like This Silent Killer Might Be Stalking You? Have you been diagnosed or pre-hypertension and hypertension? Then JOIN THE CROWD Nearly 1 in 3 adults in the United States suffer from High Blood Pressure and only 1 in 3 adults are actually aware that they have it.

Get My Free Ebook


Post a comment