Lecture #4 – June 4, 2003

Population, Screening, and Guidelines

Dr. Kunzel

Population, Screening, and Guidelines

To be familiar with the term “screening”

To recognize factors that make a disease appropriate for screening

To recognize the attributes of a good screening test

To assess oral ca as appropriate for screening

To assess oral ca exams as appropriate for screening

To consider the evidence base for oral ca screening

To consider barriers to performing oral ca detection

Screening –
a form of needs assessment

Screenings –

Procedures that sort out persons who may have a condition

from those who may not

Those appearing to have the condition are followed up

to obtain a final diagnosis

& tx as necessary

Screening –
a form of needs assessment

3 types of screening

1. Population-based

Define a population

Endeavor to screen the entire population

2. Targeted population

Aim at a “high risk” segment of the population

Gender, age, behavior, health-related characteristic

3. Opportunistic population

-- Offer screening test for unsuspected disorder

-- At time when person presents to clinician for another reason

Screening –
a form of needs assessment

What do we screen for?

Presence of disease itself

A risk factor for disease

(“do you smoke?”)

Screening –
a form of needs assessment

Dental Screenings

Cursory oral inspections to provide estimates of oral health status

Typically a dentist or dental hygienist inspects an individual’s mouth for obvious dental problems

Screening –
a form of needs assessment

Dental Screenings

Oral inspections

can be done using tongue depressors

and a simple light source, such as

flashlight

Slide 8

Screening -
a form of needs assessment

Dental Screenings

Estimates urgency of need for dental treatment

using specific criteria

Most often use the ADA’s criteria:

Apparently requires no dental tx

Requires tx but not of an urgent nature

Requires early treatment

Requires immediate treatment

Must notify examinee (or parent) of conditions that require follow-up

"2 questions need to ask"

2 questions need to ask:

What makes a disease or risk factor appropriate for screening?

What are the attributes of a good screening test?

(Frankenburg criteria)

Factors that make disease approp-riate for screening

the disease is serious (mortality, morbidity)

the disease is treatable & accepted intervention exists

early treatment is better than late treatment

the disease has a pre-clinical detectable period

the disease is prevalent in a tested population

Attributes of good screening test

high sensitivity - ability of the test to identify correctly those who have the disease

high specificity - ability of the test to identify correctly those who do not have the disease

low risk (to patient)

inexpensive or cost effective

tolerable or acceptable to the public

Attributes of good screening test

high sensitivity - ability of the test to identify correctly those who have the disease

high specificity - ability of the test to identify correctly those who do not have the disease

Screenings - must evaluate against some standard procedure for validity

sensitivity, specificity

Proportion false-positives and false negatives

important to patient follow-up and cost

(Not) attributes of good screening test

Adverse consequences of screening

Misdiagnosis - further tests must be taken if a false positive result obtained

Labeling - costs associated with telling someone have a disease

Reinforcement of bad habits among some

Attributes of good screening test

low risk (to patient)

inexpensive or cost effective

tolerable or acceptable to the public

Because screenings are applied to populations

their methods must be simple

inexpensive

require minimal training for application & interpretation

Attributes of good screening test

Positive outcomes of screening

Identification of high-risk groups for interventions –

it should be possible to differentiate those with the condition from those at borderline or without

Improved prognosis for individual patients

Reduced morbidity for cases treated early

Reduced incidence of disease

Reduced mortality

Early Oral Cancer Detection

Is oral cancer an appropriate target for screening?

Oral Ca as appropriate for screening:
Early tx better than late tx?

Rationale for oral cancer based on facts:

Malignancies are asymptomatic and localized for a period

Sometimes takes number of years to reach full invasive potential

making intervention with progression of an early lesion possible

Oral Ca as appropriate for screening:
Has a pre-clinical detectable period?

Often preceded by potentially malignant lesions/ conditions such as leukoplakia, erythroplakia, and submucous fibrosis

Major sites of occurrence readily accessible to examination

Oral Ca as appropriate for screening:
The disease is serious (mortality, morbidity)?
Early tx better than late tx?

Oral cancer has one of the lowest five-year survival rates (52%)

5-year survival rate for advanced cases is 19 %

vs. 78 % for localized lesions

When detected early, prognosis for survival better than for many other cancers

Oral Ca as appropriate for screening:
The disease is serious (mortality, morbidity)?

Morbidity:

Consequences of advanced oral cancers:

- Chronic pain

- Loss of function

- Social disfigurement

Sequelae often result in:

- cosmetic/psychological insult

- social isolation

Oral Ca as appropriate for screening:
The disease is serious (mortality, morbidity)?
The disease is prevalent in a tested population?

Mortality:

Annually approximately 30,000 Americans are diagnosed with oral cancers

1995 - 28,000 new cases

1995 - 8,400 deaths

Incidence greater than leukemia,

Hodgkin’s disease

cancer of the brain, liver, bone

thyroid gland, stomach, or cervix

Oral Ca as appropriate for screening:
The disease is prevalent in a tested population?

In the west, incidence is relatively low

Largest study group was over 23,000 adults > age 30 in MN

Mouths examined between 1957 and 1972

More than 10% of the screened had an oral lesion

But were mostly benign:

precancer encountered in 2.9%

cancer in less than 0.1%

Oral Ca as appropriate for screening:
The disease is prevalent in a tested population?

Annual incidence of oral cancer in the U.S.:

11/100,000 population

with a male/female ratio greater than 2:1

Responsible for 2% of all cancer deaths in the U.S.

"Is an oral cancer examination..."

Is an oral cancer examination a “good’ screening test?

Oral Ca exam as appropriate for screening

Principal screening test for oral cancer in asymptomatic persons

Involves visual exam of oral cavity and extraoral areas using a dental mirror

Retracting tongue with a gauze pad to view hard-to-see areas

Also digital palpation with a gloved hand for masses

Oral Ca exam as appropriate for screening
Low risk to patient?
Tolerable or acceptable to the public?
Inexpensive or cost effective?

Screening examination is inexpensive,safe

Noninvasive, quick, well tolerated

Recently published results

   large multicenter double-blind study

   suggest the potential efficacy of using the oral brush biopsy

    to increase detection of early stage oral cancer and precancerous lesions

   assuming dentists provide complete oral cancer exams on regular basis

Oral Ca exam as appropriate for screening
Inexpensive or cost effective?

Insufficient evidence of cost effectiveness

In high incidence part of the world:

Substantial proportion of suspicious lesions found (2-16% in south Asia)

But compliance of patients to attend follow up was poor

Oral Ca exam as appropriate for screening
Inexpensive or cost effective?

Screenings – patient compliance very impt

Procedures that sort out persons who may have a condition

from those who may not

Those appearing to have the condition must be followed up

to obtain a final diagnosis

& tx as necessary

Oral Ca exam as appropriate for screening
Inexpensive or cost effective?

Stronger case can be made for targeting screening to “at risk” populations

smokers

heavy drinkers > age of 40

But still problems of attendance at initial exam (patient compliance)

plus low disease prevalence

make this of uncertain utility

Oral Ca exam as appropriate for screening
Inexpensive or cost effective?

Some evidence for effectiveness of opportunistic screening

Oral Cancer Case Finding Program in Cuba

Between 1983 and 1990, 10,167,999 patients were screened when they attended clinics

only 27% with suspect lesions complied with referral

of these, 3220 potentially malignant lesions

581 squamous cell carcinomas

127 other malignancies

Oral Ca exam as appropriate for screening
Inexpensive or cost effective?

Some evidence for effectiveness of opportunistic screening

Oral Cancer Case Finding Program in Cuba (cont.)

Program was shown to be effective:

“downstaging” of cancers seen:

Stage I lesions rising from 22.8% to 48.2%

Stage II, III, and IV lesions falling from 77.2% to 51.8%

Evidence base for oral ca screening

Exam validity – its ability to differentiate:

those with the condition from those at borderline or without

Patient outcomes

Improved prognosis for individual patients

Reduced morbidity for cases treated early

Reduced incidence of disease

Reduced mortality

Evidence base for oral cancer screening

Population screening for oral ca cannot be recommended because: lack evidence of its validity:

Principal screening test in asymptomatic persons is inspection and palpation of the oral cavity

Its sensitivity is unknown

Little information on the frequency of false positives

Evidence base for oral cancer screening

Population screening for oral ca cannot be recommended because: lack evidence of its effectiveness:

No controlled trials of screening for oral cancer that include data on clinical outcomes

Consistent evidence that persons with early-stage oral cancer have a better prognosis than those diagnosed with more advanced disease

But may be due to possible effects of lead-time and length bias

Evidence base for oral cancer screening

Lead-time bias

Survival can appear to be lengthened:

when screening simply advances earlier the time of diagnosis,

lengthening the period of time between dx and death

without any true prolongation of life

Evidence base for oral cancer screening

Length bias

Tendency of screening to detect a dispro-portionate # of cases of slowly progressive disease

and to miss aggressive cases that, by virtue of rapid progression, are present in the population only briefly

Aggressive malignancies will be under-represented in the cases found

Evidence base for oc screening
Clinical Practice Guidelines

Report of the U.S. Preventive Services Task Force, 2nd ed., 1996:

Insufficient evidence to recommend for or against routine screening of asymptomatic persons for oral ca by primary care clinicians.

Although direct evidence lacking, clinicians may wish to include an exam for oral ca in the periodic health exam of persons who chew or smoke tobacco, older persons who drink regularly, anyone with suspicious symptoms

Evidence base for oc screening
Clinical Practice Guidelines

Report of the U.S. Preventive Services Task Force, 2nd ed., 1996 (cont.)

All patients, especially those at increased risk…

should be advised to receive a complete dental examination on a regular basis.

Evidence base for oc screening
Clinical Practice Guidelines

Report of the U.S. Preventive Services Task Force, 2nd ed., 1996 (cont.)

Routine oral exam by primary care clinicians

Level of evidence: III

Strength of recommendation: C

Evidence base for oc screening
Clinical Practice Guidelines

Report of the U.S. Preventive Services Task Force, 2nd ed., 1996 (cont.)

I: Evidence from at least 1 properly random- ized controlled trial

II-1: …from well-designed controlled trials w/o randomization

II-2: …from well-designed cohort or case-control analytic studies from >1 research grp

II-3: …from multiple time series w/ or w/o the intervention (dramatic results, e.g., penicillin)

III: Experts, experience, case reports

Evidence base for oc screening
Clinical Practice Guidelines

Report of the U.S. Preventive Services Task Force, 2nd ed., 1996 (cont.)

A: Good evidence to support the rec that condition be considered in periodic hlth exam

B: Fair evidence to support rec that be specifically considered

C: Insufficient evidence to rec for or against the inclusion of the condition, but rec may be made on other grounds

D: Fair evidence to support the rec that be excluded

E: Good evidence to support the rec that be

excluded

Evidence base for oc screening
Clinical Practice Guidelines

American Cancer Society

Recommends a cancer checkup

that includes oral exam every 3 years for persons over 20

and annually for those over age 40

Evidence base for oc screening
Clinical Practice Guidelines

Canadian Task Force on Periodic Health Examination

Concluded was insufficient evidence to include/ exclude screening of oral ca

in periodic health exams

of persons in the general population

but suggested

annual oral exam

for persons over 60 at risk

Evidence base for oc screening
Clinical Practice Guidelines

Logically such examinations dentists are providers of choice to perform

But is need for other health care providers to assume more responsibility

"Barriers to early oral cancer..."

Barriers to early oral cancer detection - how to confront?

To consider barriers to performing oral ca detection

A review of several studies assessing oral cancer knowledge, opinions, and practices suggests:

many MDs and DDS do not detect oral lesions in their early stages

because of inappropriate attitudes or lack of knowledge

To consider barriers to performing oral ca detection

A recent pilot survey of MD and DDS’ knowl, opinions, practices-

Found that 34% of DDS and 37% of MDs did not recognize the importance of early detection as means of reducing morbidity and mortality from these diseases

To consider barriers to performing oral ca detection

Studies report MDs do not routinely inspect their pts to identify early, suspicious oral lesions

77% of pts first diagnosed with oral ca at an advanced stage under the routine care of a MD within past 3-24 months

Another study - 94% of pts with advanced oral ca seen by a MD within 1 year of dx

To consider barriers to performing oral ca detection

DDS also remiss in early dx and referral for oral cancer

Reported that 14% of DDS performed all aspects of an intraoral exam

Studies span 3 decades, yet results unchanged

DDS missed approx twice as many asymptomatic oral ca as they found

Failed to recognize oral ca in 69% of cases presented to them


		To be familiar with the term “screening”
		To recognize factors that make a disease appropriate for screening
		To recognize the attributes of a good screening test
		To assess oral ca as appropriate for screening
		To assess oral ca exams as appropriate for screening
		To consider the evidence base for oral ca screening
		To consider barriers to performing oral ca detection


	Screenings –
	Procedures that sort out persons who may have a condition
		from those who may not
	Those appearing to have the condition are followed up
		to obtain a final diagnosis
		& tx as necessary


3 types of screening
	1. Population-based
		Define a population
		Endeavor to screen the entire population
	2. Targeted population
		Aim at a “high risk” segment of the population
		Gender, age, behavior, health-related characteristic
	3. Opportunistic population
	-- Offer screening test for unsuspected disorder
	-- At time when person presents to clinician for another reason


	What do we screen for?
		Presence of disease itself
		A risk factor for disease
	(“do you smoke?”)


	Dental Screenings
	Cursory oral inspections to provide estimates of oral health status
	Typically a dentist or dental hygienist inspects an individual’s mouth for obvious dental problems


	Dental Screenings
	Oral inspections
		can be done using tongue depressors
		and a simple light source, such as
		flashlight


	Dental Screenings
	Estimates urgency of need for dental treatment
	using specific criteria
	Most often use the ADA’s criteria:
		Apparently requires no dental tx
		Requires tx but not of an urgent nature
		Requires early treatment
		Requires immediate treatment
	Must notify examinee (or parent) of conditions that require follow-up


	2 questions need to ask:
	What makes a disease or risk factor appropriate for screening?
	What are the attributes of a good screening test?
	(Frankenburg criteria)


	the disease is serious (mortality, morbidity)
	the disease is treatable & accepted intervention exists
	early treatment is better than late treatment
	the disease has a pre-clinical detectable period
	the disease is prevalent in a tested population


	high sensitivity - ability of the test to identify correctly those who have the disease
	high specificity - ability of the test to identify correctly those who do not have the disease
	low risk (to patient)
	inexpensive or cost effective
	tolerable or acceptable to the public


	high sensitivity - ability of the test to identify correctly those who have the disease
	high specificity - ability of the test to identify correctly those who do not have the disease

	Screenings - must evaluate against some standard procedure for validity
		sensitivity, specificity
	Proportion false-positives and false negatives
		important to patient follow-up and cost


	Adverse consequences of screening
	Misdiagnosis - further tests must be taken if a false positive result obtained
	Labeling - costs associated with telling someone have a disease
	Reinforcement of bad habits among some


	Positive outcomes of screening
	Identification of high-risk groups for interventions –
		it should be possible to differentiate those with the condition from those at borderline or without
	Improved prognosis for individual patients
	Reduced morbidity for cases treated early
	Reduced incidence of disease
	Reduced mortality


	Rationale for oral cancer based on facts:
		Malignancies are asymptomatic and localized for a period
	Sometimes takes number of years to reach full invasive potential
		making intervention with progression of an early lesion possible


	Often preceded by potentially malignant lesions/ conditions such as leukoplakia, erythroplakia, and submucous fibrosis
	Major sites of occurrence readily accessible to examination


	Oral cancer has one of the lowest five-year survival rates (52%)
		5-year survival rate for advanced cases is 19 %
		vs. 78 % for localized lesions
	When detected early, prognosis for survival better than for many other cancers


	Morbidity:
	Consequences of advanced oral cancers:
	- Chronic pain
	- Loss of function
	- Social disfigurement
	Sequelae often result in:
	- cosmetic/psychological insult
	- social isolation


	Mortality:
	Annually approximately 30,000 Americans are diagnosed with oral cancers
		1995 - 28,000 new cases
		1995 - 8,400 deaths
	Incidence greater than leukemia,
		Hodgkin’s disease
		cancer of the brain, liver, bone
		thyroid gland, stomach, or cervix


In the west, incidence is relatively low
	Largest study group was over 23,000 adults > age 30 in MN
	Mouths examined between 1957 and 1972
	More than 10% of the screened had an oral lesion
	But were mostly benign:
		precancer encountered in 2.9%
		cancer in less than 0.1%


	Annual incidence of oral cancer in the U.S.:
		11/100,000 population
		with a male/female ratio greater than 2:1
	Responsible for 2% of all cancer deaths in the U.S.


	Principal screening test for oral cancer in asymptomatic persons
	Involves visual exam of oral cavity and extraoral areas using a dental mirror
	Retracting tongue with a gauze pad to view hard-to-see areas
	Also digital palpation with a gloved hand for masses


	Screening examination is inexpensive,safe
	Noninvasive, quick, well tolerated
	Recently published results
		large multicenter double-blind study
		suggest the potential efficacy of using the oral brush biopsy
		to increase detection of early stage oral cancer and precancerous lesions
		assuming dentists provide complete oral cancer exams on regular basis


	Insufficient evidence of cost effectiveness
	In high incidence part of the world:
	Substantial proportion of suspicious lesions found (2-16% in south Asia)
	But compliance of patients to attend follow up was poor


	Screenings – patient compliance very impt
	Procedures that sort out persons who may have a condition
		from those who may not
	Those appearing to have the condition must be followed up
		to obtain a final diagnosis
		& tx as necessary


	Stronger case can be made for targeting screening to “at risk” populations
	smokers
	heavy drinkers > age of 40

	But still problems of attendance at initial exam (patient compliance)
	plus low disease prevalence
	make this of uncertain utility


	Some evidence for effectiveness of opportunistic screening
	Oral Cancer Case Finding Program in Cuba
		Between 1983 and 1990, 10,167,999 patients were screened when they attended clinics
		only 27% with suspect lesions complied with referral
		of these, 3220 potentially malignant lesions
		581 squamous cell carcinomas
		127 other malignancies


	Some evidence for effectiveness of opportunistic screening
	Oral Cancer Case Finding Program in Cuba (cont.)
	Program was shown to be effective:
		“downstaging” of cancers seen:
		Stage I lesions rising from 22.8% to 48.2%
		Stage II, III, and IV lesions falling from 77.2% to 51.8%


	Exam validity – its ability to differentiate:
		those with the condition from those at borderline or without
	Patient outcomes
	Improved prognosis for individual patients
	Reduced morbidity for cases treated early
	Reduced incidence of disease
	Reduced mortality


	Population screening for oral ca cannot be recommended because: lack evidence of its validity:
		Principal screening test in asymptomatic persons is inspection and palpation of the oral cavity
		Its sensitivity is unknown
		Little information on the frequency of false positives


	Population screening for oral ca cannot be recommended because: lack evidence of its effectiveness:
	No controlled trials of screening for oral cancer that include data on clinical outcomes
	Consistent evidence that persons with early-stage oral cancer have a better prognosis than those diagnosed with more advanced disease
		But may be due to possible effects of lead-time and length bias