Sbu • Statens beredning för medicinsk utvärdering The Swedish Council on Technology Assessment in Health Care

variables

Validating 

criteria

Statist-

ical met-

hods

Results*

Sensitivity and 

specificity

Study 

quality and 

relevance

Comments

Petti et al

2000 [77]

Italy

3 primary 

schools  

in Rome

caries-free 

only included 

(39% of total

sample)

n: 314

Age: 6–7

Primary 

and per-

manent

Mean

dft: 1.6 

DMFT: 0.6

Increment

Mean new

dmft: 0.5,

Mean new

DMFT: 0.2

2

3

WHO 

1987

BW

3 exam-

iners

Yes

Repeated salivary MS

0 vs 1–3, 0–1 vs 2–3,

0–2 vs 3 positive

tests

Fluoride intake, 

sucrose intake 

plaque

≥1 new dft

or DFT

If exfoliated

at follow-up:

counted as 

caries

Only 

caries-free 

at baseline 

included

Se, Sp, 

PPV, NPV

Modest improve-

ment of MS multiple

tests

0 vs 1–3 positive

tests

Se: 50%; Sp: 80%;

PPV: 0.59; NPV:

0.74

Low F, high plaque

Se: 57%; Sp: 94%

Low

Only caries-

free included. 

Bias from 

study design

Pienihäkkinen

1987 [44]

Finland

Institutional-

ised Hungarian 

children

1. Xylitol 

group

2. F and con-

trol group

n: 276

Age: 7–12

Permanent

Mean

D

en

MFS:

1.8

Increment

Mean new

D

en

MFS:

2.0

2

30

Analysed

WHO

No BW 

(fibre

optics)

7 exam-

iners at 

start,  

3 there-

after

Yes

≥1 incipient lesion 

or buccal or lingual 

surfaces

LB>10

5

CFU/ml

saliva

Yeasts ≥5 CFU/ml

Saliva salivary buffer 

capacity: pH ≤5 and 

combinations of 

these

Age ≤9:

2–3 new

DMFS

Age ≥10:

4–5 new

DMFS

Proportion 

high risk: 

Median value

(value not

stated)

Se, Sp, log 

regression

Incipient lesions (inc) 

Se: 52%; Sp: 87%

Combinations

Inc and/or LB + 

yeasts  

Se: 67%; Sp: 76%

Inc and/ or LB + 

yeasts

Se: 70%; Sp: 71%

Salivary buffering  

Se: 63%; Sp: 69

Accuracy max 0.77

Low

Very caries-

active 

children;

open cavities 

common; not

applicable to 

populations 

of today

Pienihäkkinen

1987 [45]

Finland

Institutional-

ised

Hungarian 

children 

(n=350)

1. Xylitol group

2. F and con-

trol group

n: 276

Age: 7–12

Permanent

Mean

D

en

MFS:

1.8

Increment

Mean new

D

en

MFS:

2.0

2

19

WHO

No BW 

(fibre

optics)

7 exam-

iners at 

start,  

3 there-

after

Yes

Salivary buffering;  

2 levels:  

pH ≤4 and pH ≤5

≥3 new

DMFS

Proportion 

high risk: 

Varying

Se, Sp

Best for pH ≤5 

at age ≤9 yrs

Se: 58% girls;  

52% boys

Sp: 64% girls;  

63% boys

Same 

material as 

Pienihäkkinen 

1987 [44].

Important

confounders 

not included

The table continues on the next page

K A R I E S   –   d I A G N O S T I K ,   R I S K B E d ö M N I N G   O C H   I C K E - I N vA S I v   B E H A N d L I N G

264

Table 5.4 continued

Author

Year,  

reference

Country

Sample, n

Age (years) 

at start

Teeth 

tested,

Caries 

prevalen-

ce at start

Obs 

time 

(yrs)

Drop-out

Explained

Diag-

nostic 

criteria

Exam-

iner (n) 

Relia-

bility

Predictor  

variables

Validating 

criteria

Statist-

ical met-

hods

Results*

Sensitivity and 

specificity

Study 

quality and 

relevance

Comments

Raadal et al

1992 [30]

Norway

All children 

born 1978–

1979 attending 

a public dental 

clinic in a town

in Norway

n: 192

Age: 6

Occlusal 

fissures 1st 

permanent 

molars

Baseline: 

Mean dmft

3.4

Increment:

Not stated

2

Not 

stated

Möller

1966

BW

3 exam-

iners

No

Baseline dmft  

(from 0–11)

≥1 molar 

with initial

or dentine 

caries or ≥1 

molar with

dentine  

caries

Proportion 

high risk: 

27–46% (all

proportions 

reported)

Se, Sp, 

PPV, NPV

Best combined values

for ≥3 dmft and ≥1

molar with dentine

caries with proportion

high risk: 46%

Se: 72%; Sp: 67%;

PPV: 0.52; NPV:

0.83

For proportion high

risk: 27% ≥5 dmft

and ≥1 molar with

dentine caries)

Se: 44%; Sp: 82%;

PPV: 0.55; NPV:

0.74

Low

Only 1 

predictor. 

Incom-

plete data 

reporting. 

Possible bias 

from treat-

ment

Seppä et al

1988 [78]

Finland

Selected from 

a clinical trial 

on F-varnish

n: 124

Age: 11–13

(mean 12)

Permanent

Mean

DMFS 17.7

Increment:

Mean new

DMFS 14.1

5

34

Explained

Möller &

Poulsen 

1973

BW

2 exam-

iners

Yes

Baseline DS, FS,

initial lesions

DFS: Not

stated

Proportion 

high risk: 

30%

Se, Sp, 

PPV, NPV,

accuracy, 

OR

DFS with (without)

initial lesions

Se: 49% (51%);  

Sp: 76% (78%);  

PPV: 0.45 (0.47);

NPV: 0.79 (0.80)

Initial caries added

little to the predic-

tive power

Low

Retrospec-

tive bias from 

treatment 

(F varnish).

incom-

plete data 

reporting

Seppä et al

1989 [79]

Finland

High risk 

individuals 

in Kuopio +

random sample 

of low risk

n: 512

Age: 6

Primary 

and per-

manent

dmfs: 0.49

DMFS:

0.37

Increment:

Mean 4.4

new DMFS

7, 

step-

wise

pre-

dic-

tion

10

Möller &

Poulsen 

1973 

BW at 

the end

3 exam-

iners 

“trained 

research 

tem”

No 

reliability 

test

Caries experience 

each year from age 

6 to 12

DMFS >14

Proportion 

high risk: 

10–52%

Se, Sp  

for diff

Propor-

tions 

believed 

to be high 

risk

Se lowest at age 6:

28% (Sp: 92%); and

highest at age 9:  

Se: 68% (Sp: 85%)

For 30% high risk

proportion, age 9 

Se: 61%; Sp: 81%;

PPV: 0.53; NPV:

0.85

Low

Retrospec-

tive. Bias 

from treat-

ment

265

K A P I T E L   5   •   r I s K b E d ö M n I n g

Table 5.4 continued

Author

Year,  

reference

Country

Sample, n

Age (years) 

at start

Teeth 

tested,

Caries 

prevalen-

ce at start

Obs 

time 

(yrs)

Drop-out

Explained

Diag-

nostic 

criteria

Exam-

iner (n) 

Relia-

bility

Predictor  

variables

Validating 

criteria

Statist-

ical met-

hods

Results*

Sensitivity and 

specificity

Study 

quality and 

relevance

Comments

Raadal et al

1992 [30]

Norway

All children 

born 1978–

1979 attending 

a public dental 

clinic in a town

in Norway

n: 192

Age: 6

Occlusal 

fissures 1st 

permanent 

molars

Baseline: 

Mean dmft

3.4

Increment:

Not stated

2

Not 

stated

Möller

1966

BW

3 exam-

iners

No

Baseline dmft  

(from 0–11)

≥1 molar 

with initial

or dentine 

caries or ≥1 

molar with

dentine  

caries

Proportion 

high risk: 

27–46% (all

proportions 

reported)

Se, Sp, 

PPV, NPV

Best combined values

for ≥3 dmft and ≥1

molar with dentine

caries with proportion

high risk: 46%

Se: 72%; Sp: 67%;

PPV: 0.52; NPV:

0.83

For proportion high

risk: 27% ≥5 dmft

and ≥1 molar with

dentine caries)

Se: 44%; Sp: 82%;

PPV: 0.55; NPV:

0.74

Low

Only 1 

predictor. 

Incom-

plete data 

reporting. 

Possible bias 

from treat-

ment

Seppä et al

1988 [78]

Finland

Selected from 

a clinical trial 

on F-varnish

n: 124

Age: 11–13

(mean 12)

Permanent

Mean

DMFS 17.7

Increment:

Mean new

DMFS 14.1

5

34

Explained

Möller &

Poulsen 

1973

BW

2 exam-

iners

Yes

Baseline DS, FS,

initial lesions

DFS: Not

stated

Proportion 

high risk: 

30%

Se, Sp, 

PPV, NPV,

accuracy, 

OR

DFS with (without)

initial lesions

Se: 49% (51%);  

Sp: 76% (78%);  

PPV: 0.45 (0.47);

NPV: 0.79 (0.80)

Initial caries added

little to the predic-

tive power

Low

Retrospec-

tive bias from 

treatment 

(F varnish).

incom-

plete data 

reporting

Seppä et al

1989 [79]

Finland

High risk 

individuals 

in Kuopio +

random sample 

of low risk

n: 512

Age: 6

Primary 

and per-

manent

dmfs: 0.49

DMFS:

0.37

Increment:

Mean 4.4

new DMFS

7, 

step-

wise

pre-

dic-

tion

10

Möller &

Poulsen 

1973 

BW at 

the end

3 exam-

iners 

“trained 

research 

tem”

No 

reliability 

test

Caries experience 

each year from age 

6 to 12

DMFS >14

Proportion 

high risk: 

10–52%

Se, Sp  

for diff

Propor-

tions 

believed 

to be high 

risk

Se lowest at age 6:

28% (Sp: 92%); and

highest at age 9:  

Se: 68% (Sp: 85%)

For 30% high risk

proportion, age 9 

Se: 61%; Sp: 81%;

PPV: 0.53; NPV:

0.85

Low

Retrospec-

tive. Bias 

from treat-

ment

The table continues on the next page

K A R I E S   –   d I A G N O S T I K ,   R I S K B E d ö M N I N G   O C H   I C K E - I N vA S I v   B E H A N d L I N G

266

Table 5.4 continued

Author

Year,  

reference

Country

Sample, n

Age (years) 

at start

Teeth 

tested,

Caries 

preval-

ence at 

start

Obs 

time 

(yrs)

Drop-out

Explained

Diag-

nostic 

criteria

Exam-

iner (n) 

Relia-

bility

Predictor  

variables

Validating 

criteria

Statist-

ical met-

hods

Results*

Sensitivity and 

specificity

Study 

quality and 

relevance

Comments

Splieth et al

1999 [80]

Germany

Sample 

selection not 

described

n: 169

Age: 6

Permanent, 

1st molars

all caries-

free

Mean  

dmfs: 9.9

Mean

DMFS: 0.35

Increment:

Mean new

DMFS 0.34

2

11

WHO 

1987 

modi-

fied and 

Martha-

ler 1990

No BW

1 exam-

iner

Not 

stated

MS in fissures  

and in plaque

≥1 DMFS in

fissures 1st 

permanent 

molars

Proportion 

high risk: 

28%

Se, Sp

MS low vs MS high

Se: 50%; Sp: 82%;

PPV: 0.29; NPV:

0.92

Low

Bias from 

treatment: 

24% fissure

sealed. No 

reprodu- 

cibility test

Stecksén-

Blicks

1985 [81]

Sweden

Randomly

selected from 

target popu-

lation Umeå, 

Sweden

n: 88+91

Age: 8 or 14

Primary 

and per-

manent

Mean,  

age 8:

dmfs=7.1;

DMFS=3.3

Increment:

Not stated

1

23

Koch 

1967

BW

No of 

exam-

iners 

unclear

Not 

stated

MS, LB in saliva

≥3 new

lesions 

(including

initial)

≥10

5

LB

≥10

5

MS

Proportion 

high risk: 

Around 20%

Se, Sp, 

PPV, NPV

8 yr-olds

MS alone

Se: 67%; Sp: 74%

LB alone

Se: 44%; Sp: 86%

MS and LD 

Se: 72%; Sp: 67%

Low

Too short 

follow-up for

permanent 

teeth. Incom-

plete data 

reporting

Swenson et al

1976 [82]

USA

Children from 

public elemen-

tary schools 

in Roseville,

Minnesota

n: 575

Age: 6–9

Primary 

and per-

manent

Mean  

dfs: 4.4 

Mean

DMFS:

0.34

Increment:

Mean new

dfs 1.0

Mean new

DMFS 1.8

2

26

Explained

Proceed-

ings 

1972

No BW

Not 

stated

MS, sanguis, mitior,

salivarius, LB >1%

MS in pooled plaque

was used as cut-off

>0 new DFS

Proportion 

high risk: 

49%

Se, Sp, 

PPV, NPV

Only MS was stat-

istically significantly 

associated with

caries increment

Se: 55%; Sp: 76%;

PPV: 0.9; NPV: 0.29

Low

Incom-

plete data 

reporting. 

Bias from 

diagnosis and 

treatment 

possible

267

K A P I T E L   5   •   r I s K b E d ö M n I n g

Table 5.4 continued

Author

Year,  

reference

Country

Sample, n

Age (years) 

at start

Teeth 

tested,

Caries 

preval-

ence at 

start

Obs 

time 

(yrs)

Drop-out

Explained

Diag-

nostic 

criteria

Exam-

iner (n) 

Relia-

bility

Predictor  

variables

Validating 

criteria

Statist-

ical met-

hods

Results*

Sensitivity and 

specificity

Study 

quality and 

relevance

Comments

Splieth et al

1999 [80]

Germany

Sample 

selection not 

described

n: 169

Age: 6

Permanent, 

1st molars

all caries-

free

Mean  

dmfs: 9.9

Mean

DMFS: 0.35

Increment:

Mean new

DMFS 0.34

2

11

WHO 

1987 

modi-

fied and 

Martha-

ler 1990

No BW

1 exam-

iner

Not 

stated

MS in fissures  

and in plaque

≥1 DMFS in

fissures 1st 

permanent 

molars

Proportion 

high risk: 

28%

Se, Sp

MS low vs MS high

Se: 50%; Sp: 82%;

PPV: 0.29; NPV:

0.92

Low

Bias from 

treatment: 

24% fissure

sealed. No 

reprodu- 

cibility test

Stecksén-

Blicks

1985 [81]

Sweden

Randomly

selected from 

target popu-

lation Umeå, 

Sweden

n: 88+91

Age: 8 or 14

Primary 

and per-

manent

Mean,  

age 8:

dmfs=7.1;

DMFS=3.3

Increment:

Not stated

1

23

Koch 

1967

BW

No of 

exam-

iners 

unclear

Not 

stated

MS, LB in saliva

≥3 new

lesions 

(including

initial)

≥10

5

LB

≥10

5

MS

Proportion 

high risk: 

Around 20%

Se, Sp, 

PPV, NPV

8 yr-olds

MS alone

Se: 67%; Sp: 74%

LB alone

Se: 44%; Sp: 86%

MS and LD 

Se: 72%; Sp: 67%

Low

Too short 

follow-up for

permanent 

teeth. Incom-

plete data 

reporting

Swenson et al

1976 [82]

USA

Children from 

public elemen-

tary schools 

in Roseville,

Minnesota

n: 575

Age: 6–9

Primary 

and per-

manent

Mean  

dfs: 4.4 

Mean

DMFS:

0.34

Increment:

Mean new

dfs 1.0

Mean new

DMFS 1.8

2

26

Explained

Proceed-

ings 

1972

No BW

Not 

stated

MS, sanguis, mitior,

salivarius, LB >1%

MS in pooled plaque

was used as cut-off

>0 new DFS

Proportion 

high risk: 

49%

Se, Sp, 

PPV, NPV

Only MS was stat-

istically significantly 

associated with

caries increment

Se: 55%; Sp: 76%;

PPV: 0.9; NPV: 0.29

Low

Incom-

plete data 

reporting. 

Bias from 

diagnosis and 

treatment 

possible

The table continues on the next page

K A R I E S   –   d I A G N O S T I K ,   R I S K B E d ö M N I N G   O C H   I C K E - I N vA S I v   B E H A N d L I N G

268