Reference Values for Respiratory Rate in the First 3 Years of Life

(1)

Reference

Values

for

Respiratory

Rate

in the

First

3 Years

of Life

Franca Rusconi, MD*; Manuela Castagneto, MD*; LUigi Gagliardi, MDX; Gualtiero Leo, MD;

Antonio Pellegatta, MDIJ; Norberto Porta, MD*; Sultana Razon, MDII; and Mario Braga, MD**

ABSTRACT. Background. Raised respiratory rate is a

useful sign to diagnose lower respiratory infections in childhood. However, the normal range for respiratory rate has not been defined in a proper, large sample.

Objective. To assess the respiratory rate in a large number of infants and young children in order to

con-struct percentile curves by age; to determine the

repeat-ability of the assessment using a stethoscope and com-pare it with observation.

Methods. Respiratory rate was recorded for 1 minute with a stethoscope in 618 infants and children, aged 15 days to 3 years old, without respiratory infections or any

other severe disease when awake and calm and when

asleep. In 50 subjects we compared respiratory rate taken 30 to 60 minutes apart to assess repeatability, and in 50 others we compared simultaneous counts obtained by stethoscope versus observation.

Results. Repeatability was good as the standard de-viation of differences was 2.5 breaths/minute in awake and 1.7 breaths/minute in asleep children. Respiratory rate obtained with a stethoscope was systematically higher than that obtained by observation (mean differ-ence 2.6 breaths/minute in awake and 1.8 breaths/minute in asleep children; P .015 and P < .001, respectively). A

decrease in respiratory rate with age was seen for both

states, and it was faster in the first few months of life when also a greater dispersion of values was observed. A

second degree polynomial curve accurately fitted the

data. Reference percentile values were developed from these data.

Conclusions. The repeatability of respiratory rate

measured with a stethoscope was good. Percentile curves

would be particularly helpful in the first months of life when the decline in respiratory rate is very rapid and prevents to use cut off values for defining “normality.”

Pediatrics 199494:350-355; respiratory rate, age-related

centiles, reference values, repeatability.

ABBREVIATION. RR, respiratory rate.

In many studies tachypnea has been found to be

the most reliable clinical sign that a child has

pneu-From the *First and Second Departments of Pediatrics, University of Mi-lano; Department of Neonatology and NICU, Ospedale Niguarda, Milano; Departments of Pediatrics of §Ospedale Regina Elena and ilOspedale

S. Carlo, Milano and lOspedale di Busto Arsizio; **Epideimology Unit, Ospedale Maggiore Policlinico, Milano, Italy.

These data were presented in part at the Annual Meeting of the American Thoradc Society-American Lung Association International Conference,

San Francisco, May 16 through 19, 1993.

Received for publication Sep 27, 1993; accepted Jan 31, 1994.

Reprint requests to (FR.) Seconda Clinica Pediatrica, Universit#{224} di Milano, Via Commenda 9, 20122 Milano, Italy.

monia rather than an upper respiratory tract

infec-tion.’4 Although most of these studies were carried

out in developing countries to help primary health

workers to manage respiratory infections correctly,

also in developed countries many infants and young

children with respiratory infections are seen in

out-patient settings where roentgenographic facilities are not immediately available. In these situations a more

precise definition of tachypnea and percentile curves

for reference values would be particularly useful.

The only centile curves available do not cover the

first years of life in detail, and were obtained 40 years

ago by repeated measurements in few subjects

re-gardless whether they were awake or asleep.5

Fur-thermore, these infants and children lived at high

altitude (Denver, CO), and the lower oxygen partial

pressure could therefore have influenced the results.6

Surprisingly only a few other studies have dealt

with the problem of defining the “normal”

respira-tory rate. In a community-based study7 the

respira-tory rate was counted for 60 seconds in a very large

number (>6000 examinations) of infants younger

than 12 months in different states (sleeping, awake

and calm, awake and not calm). Also this study was

done at high altitude (Lima, Peru), and, moreover,

the infants had a history of cough and therefore

could have had a respiratory tract disease. The other

two studies available on a large number of infants

refer only to subjects under 6 months.8’9 Richards et

al8 reported the distribution of mean respiratory rate

values recorded during sequential 22-hour electronic

monitoring: this method of counting is impractical to

use in clinical settings and the data cannot therefore

be used as reference values by clinicians. In Morley

et al’s study9 the mean respiratory rate was assessed

by stethoscope or by a hand on the chest during three

15-second intervals in asleep and awake infants; this

short time of counting could have introduced some

error in the final evaluation of respiratory rate. The

mean values found by the authors are higher than

those previously described and do not vary with

postnatal age.

These discrepancies in methodologies,

popula-tions studied, and results, and the lack of age-related reference values for the first years of life led us to

perform this study. The aim was to record the

respi-ratory rate in infants and children aged 15 days to 3

years with neither severe diseases nor clinical signs

of respiratory infections in order to: 1) determine the

repeatability of recordings 30 to 60 minutes apart

by stethoscope, which is a very common method to

(2)

compare respiratory rate counts obtained by

stethoscope and by observation; and 3) construct

reference centile curves by age in a large number

of subjects.

SUBJECTS AND METHODS

Six hundred eighteen infants and children, 336 males and 282

females, aged 15 days to 3 years, were assessed during 1 year. Distribution in this age range is shown in Table 1. The subjects were seen evenly across the seasons by four investigators. Each

investigator assessed a minimum of 100 subjects. Of the 618 sub-jects assessed, 309 were in good health and were seen in day care centers and 309 were seen in hospital as inpatients oroutpatients.

The main diagnoses for these patients were gastrointestinal and urologic disorders, skin diseases, and orthopedic problems.

Exclusion criteria were the presence of any chronic or severe

illness, dehydration, or a history of fever or respiratory findings suggesting a respiratory infection in the previous 2 weeks.

In each child the respiratory rate was recorded by the same assessor both when the child was awake and calm and when he was in quiet sleep according to behavioral criteria (sleep not

associated with any spontaneous movement, induding no eye movements or vocalizations). Infants and children who were un-settled or crying did not have their respiratory rate measured. In awake children older than I year of age respiratory rate was

assessed when they were playing quietly. For each state respira-tory rate was assessed by listening to breath sounds for 60 seconds with a stethoscope put gently on the baby’s naked chest.

In 50 subjects evenly distributed across the ages (10 for each of the following dasses of age: <2 months, 2 to <6, 6 to <12, 12 to <24, and 24 to 36 months) the respiratory rate was recorded twice, by the same assessor, 30 to 60 minutes apart, listening to breath sounds for 60 seconds with a stethoscope, when the child was awake and calm and when he was sleeping.

In 50 subjects evenly distributed across the ages (same as

above) the respiratory rate in the two different states (awake and calm and sleeping) was recorded simultaneously by two assessors, one listening to breath sounds and the other counting the

respi-ratory rate by observation of the abdominal and chest wall

move-ments. To avoid as far as possible an operator effect, the two assessors exchanged roles (observer-listener) randomly.

STATISTICS

Repeatability Analysis and Comparison Between

Observed and Stethoscope Counts

The repeatability of respiratory rate measurement

by stethoscope was evaluated by calculating for each

pair of recordings the difference between the first

and the second measurement (first minus second

measurement).’#{176} The standard deviation of these dif-ferences is an index of the precision of the method.

The SD of the differences was used to calculate the

95% coefficient of repeatability’0 (“the value

be-low which the differences between two single test

results may be expected to lie with a specified

probability”).”

The agreement between respiratory rate counts

obtained by stethoscope and those obtained by

ob-servation was assessed using the same technique, ie,

by calculating the difference in respiratory rate

counts (stethoscope minus observation).’#{176} The mean

of the difference between the two counts is an

esti-mate of the systematic difference (bias) between the

methods.

State of the Child and Influence of Potential Confounding Factors

The influence of the state of the child (awake and

calm or asleep) and of potential confounding factors

(gender, season of the year, and examination in

hos-pital or in day care centers) on respiratory rate

counts was assessed with an analysis of variance

after grouping the subjects in the following classes of

ages: <2, 2 to <6, 6 to <12, 12 to <24, 24 to <30, and 30

to 36 months. All these calculations were done with

the statistical package SPSS/PC+.

Construction of Centile Curves

Two problems were encountered in the

construc-tion of centile curves from these data; first, the skew-ness of the distribution of respiratory rate at different ages was highly variable; and second, the dispersion (SD) of the values at various ages was also different.

No single data transformation could resolve these

problems and render the data identically and

nor-mally distributed across all age values, so as to

permit the use of standard techniques. Log

transfor-mation however led to equal variances,’2 and

conse-quently was used before the subsequent analysis,

which was carried out with the nonparametric

tech-nique described by Healy et al.’3 Briefly, an ideal box

around the first k (in our case k = 50) observations

sorted according to age values is selected, and the

least squares regression of respiratory rate on age is

then calculated. The centiles of the observed

residu-als from the fitted line are then calculated and

as-signed to the median of the age values of the box.

The box is then moved I point ahead (removing the

first point, and including the 51st one), and the

pro-cedure is repeated until all data are covered and raw

centiles are obtained.

With this method the first half of the first box and

the second haM of the last box do not contribute

points to raw centiles. To avoid losing this

informa-tion we oversampled subjects at younger ages, and,

at the other extreme, we studied also 44 children

older than 36 months of age. The raw centiles

ob-tamed in the first step of the procedure are very

irregular, and need to be smoothed. This is done by

TABLE. Mean, median ,and SD of respira tory rate acco rding to waking or sl eeping in 61 8 infants and children grouped in c lasses of age

Age (Months) Number Awake Subjects Asleep Subjects

Mean Median SD Mean Median SD

<2 2 to <6

6 to <12

12 to <18 18 to <24 24 to <30 30 to 36

104 106 126 77 65 79 61 48.0 44.1 39.1 34.5 32.0 30.0 27.1 47 42 38 34 32 30 28 9.1 9.9 8.5 5.8 4.8 6.2 4.1 39.8 33.4 29.6 27.2 25.3 23.1 21.5 39 32 28 26 24 23 21 8.7 7.0 7.0 5.6 4.6 4.6 3.7

at Viet Nam:AAP Sponsored on September 1, 2020 www.aappublications.org/news

(3)

C E UI .C 15 .0 C U C 5, 5,

) 10 30 50 70

C E UI .5 a, .0 C 0 (5 > a, UI .0 a, 0. 0 .C 20 a, fl c io c S

C S S

!

S

Average RR by the two counts (breaths/mm) Fig 2. Differences between respiratory rate (RR) counts simulta-neously determined by stethoscope and by observation in relation to their mean in awake (A) and asleep (B) subjects.

0 20 40 60 80

Average RR by the two counts (breaths/mmn)

Fig 1. Differences in respiratory rate (RR) counts determined by

stethoscope at 30 to 60 minutes interval in relation to their mean in

awake (A) and asleep (B) subjects.

fitting interrelated polynomial functions to the

ob-served centiles (details of this second step are

out-lined in the Appendix), which guarantees as much

flexibifity as possible without losing information as

the centile curves are constrained to vary smoothly

with age and also their relation varies smoothly.

Previous knowledge and careful data inspection

helps to determine the degree of constraint that

should be imposed on the individual curves.

Separate centile curves were fitted to the data

ob-tamed in awake and asleep states. A pilot package

(GROSTAT II) implementing the method was used

for the centile construction.14

RESULTS

Repeatability of Respiratory Rate Counts

In Fig 1 the differences between respiratory rate

counts determined by stethoscope at 30-60 minutes

interval are plotted against their mean count. The SD

of the differences between the two counts was 2.5

and 1.7 breaths/minute, respectively, when the

in-fants and children were awake and calm and when

they were asleep (yielding 95% repeatability

coeffi-cients of 4.9 and 3.3 breaths, respectively). The size of

the differences and the mean respiratory rate value

were not correlated (r = .013 and .035 in awake and

in asleep subjects, respectively), ie, the respiratory rate value did not seem to influence the repeatability of the observation.

‘U

A S

5 .

0

55 5 5

555 5 55. 5 S 5 5 55 5 5

S S S

S S S S S S S S 555 S S S S -5 -10 0 20 ! S

40 60 81

5

0

-5

-1(

B

S 55 5

55

S 555 5 5

5

S S S S S S

S S S

S #{149}S S

55

I I

Comparison Between Respiratory Rate Obtained by

Stethoscope and by Observation

Both in awake and in asleep states respiratory rate counts obtained by stethoscope were systematically

higher than those obtained by observation (Fig 2).

The mean difference (mean bias) was 2.6 and I .8

breaths/minute, respectively, in awake and asleep

states (P = .015 and P < .001). The dispersion was

wider in infants and children when awake than

when asleep (SD of the differences 7.4 vs. 2.3 breaths,

respectively). A tendency toward a greater

disper-sion at higher respiratory rate was noted.

State of the Child and Influence of Potential

Confounding Factors

At all ages mean respiratory rate counts were

sig-nificantly higher (P < .001, paired Student’s t test

with Bonferroni correction for multiple comparisons)

in infants and children when awake and calm than

when asleep (Table 1).

The effect of potentially confounding factors

(gen-der, season of the year, or being seen in a day care

center or in a hospital) was then assessed separately

in awake and in asleep states with an analysis of

variance. None of the factors examined significantly

influenced the respiratory rate count in either state so

that the data were considered together in the

con-struction of centile curves.

Postnatal Age and Construction of Centile Curves

A decrease in respiratory rate with increasing age

was seen both in awake and in asleep infants and

(4)

80 70 60 50 40 30

C E

UI

.C

(5 0 .0

Age (months)

27 30 33 36

C

E .C

(5

a,

.0

60

50 40 30 20

10

0 3 6 9 12 15 18 21 24 27 30 33 36

Age (months)

Fig 5. 5th, 10th, 25th, 50th, 75th, 90th, and 95th smoothed centiles by age in awake subjects (RR = respiratory rate).

70 60

C

.5o

. 40

.0

.- 30

20

0 3 6 9 12 15 18 21 24 27 30 33 36

Age (months)

Fig 6. 5th, 10th, 25th, 50th, 75th, 90th, and 95th smoothed centiles by age in asleep subjects (RR = respiratory rate).

S.

I’..

U:

40

.0

a: a:

20

10

0 3 6 9 12 15 18 21 24 27 30 33 36

Age (months)

Fig 3. Scatter diagram of respiratory rate (RR) counts by age in awake subjects.

children (Table 1, Figs 3 and 4). The pattern of

respi-ratory rate decline with age was similar in both states

and was much faster in the first few months of life.

To be able to describe with greater accuracy and

precision the steeper part of the curves, where the

observed fall in respiratory rate was faster, we

sam-pled more babies in the first year of life and, in

particular, in the first few months. The dispersion of

observed data was wider at young ages: the SDs of

the observed respiratory rate counts fell from 9.1

breaths/minute in subjects <2 months to 4.1 breaths/

minute in subjects 30 to 36 months old in the awake

state, and from 8.7 to 3.7 breaths/minute in the

asleep state (Table 1).

To construct centile curves the data were first

log-transformed and then a second degree (quadratic)

polynomial curve was used. This allowed an

excel-lent fitting to observed data with very simple

equa-tions. In Figs 5 and 6 the centile curves for awake and

asleep infants and children are shown. The equations

describing the 95th centile are:

y = 10(1821 - O.0125995x + O.00013401x2),

in the awake state, and

y = 10(1 - 0.0139928x + 0.00017646x)

in the asleep state, where y = respiratory rate, and

x = age in months.

The Appendix gives the general equations describing

the centile curves for both states.

Fig 4. Scatter diagram of respiratory rate (RR) counts by age in

asleep subjects.

70

DISCUSSION

The difficulty in defining the “normal” respiratory

rate in infants and young children prompted us to

perform this study. Differences between the reported

respiratory rates in healthy infants and children may

be due to various factors as extensively discussed in

a recent review by Berman et al,7 including the

num-ber of infants studied, the period of time counting, the method of counting, and the state of the infant.

Because of the natural variability of the respiratory

rate and the presence of periodic breathing in infants

<

6

months of age,8’15 a 60-second count has been

recommended by the World Health Organization.

One minute’s counting either at a stretch or in two

blocks of 30-second intervals has recently been

con-firmed to be the best time period of counting by

Simoes et al’6 in an ad hoc study.

Many techniques have been used to count the

respiratory rate in infants and children including

continuous electronic monitoring from an abdominal

or thoracic wall movement detector, observation of

the abdominal and chest wall movements, listening

to the breath sounds with a stethoscope, and putting

a hand on the baby’s chest. Recording respiratory

rate by electronic monitoring is obviously impracti-cal in clinical practice. Moreover, with this method,

particularly if the infant is asleep, movements could

distort respiratory tracings or superficial breaths

could be missed.8’17 These problems could also affect

the observation counting method.7”8

(5)

As there is no counting method that can be

considered a gold standard we chose to use the

stethoscope because, like Morley,’8 we found it

very difficult to count respiratory rate by

observa-tion in very active, lively infants.

Use of the stethoscope yielded good repeatability

and low variability of the respiratory rate over a

short period of time (30 to 60 minutes) both in awake

and in sleeping infants and children. This counting

method probably allows detection of even small

breaths that are missed at observation because of

chest/abdominal movements. In fact, in subjects in

whom simultaneous (stethoscope and observation)

counts were performed we found that the respiratory

rate recorded by stethoscope was systematically

higher than that recorded by observation.

In our study, as in previous ones,9”5”7 respiratory

rate counts were higher in awake than in sleeping

subjects and decreased with increasing age. Our

mean respiratory rate values in the first 6 months of

life were comparable to those reported in studies in

which continuous electronic monitoring was used

for prolonged time periods to record respiratory

rate5”7 but lower than those reported by Morley

et al.9

The variability of respiratory rate among subjects

was larger in the first few months of life in our series.

Analysis of the causes is beyond the scope of the

study, but there are many possible biological

expla-nations. In the first months of life several events

occur including maturation of the neurologic control

of breathing (eg, the weakening and disappearance

of reflexes such as the Hering-Breuer reflex’9), the

change of the hemoglobin dissociation curve due to

the shift from fetal to adult hemoglobin, with

differ-ences in oxygen delivery to tissues, changes of the

mechanical properties of the respiratory system (eg,

changes in lung and chest wall compliance and lung

volumes1921), all of which probably affect

respira-tion. The degree and speed of these phenomena

probably vary greatly from one baby to another, and

result in greater overall variability.

Pediatricians are accustomed to use age-related

centiles when evaluating findings that change with

age. However, in the evaluation of respiratory rate

age-related cutoff values are used more frequently

than centile curves. The graphic detail of the only

centile curves available5 is poor, particularly for the

first years of age. This constitutes an obstacle to their

clinical use, in addition to the problems already

dis-cussed in the collection of the raw data used to

construct these curves. Moreover, since the

smooth-ing of the data was carried out graphically, no

equa-tions are provided and it is impossible to redraw the

figures. On the other hand, cutoff values and

tabu-lations of descriptive statistics such as means and

SDs in classes of age are very inefficient ways of

presenting data when age-related changes are rapid,

and are potentially misleading unless very narrow

classes of age are considered.

The curves presented here allow the decline in the

respiratory rate to be followed with greater detail

and accuracy, especially in the first months of life.

The use of the distribution-free technique developed

by Healy et al’3 permitted us to construct accurate

centile curves without assumptions on the shape of

the distribution at each age and considering only the

general trend of respiratory rate. A second degree

polynomial curve was sufficient to describe the data

accurately both in awake and in asleep infants and

children. Only five parameters were required to fully

construct the centiles. Thus, the danger of overfitting the data was avoided.

In conclusion, we hope that the curves presented

in this study will be helpful, particularly in

evaluat-ing babies in the first months of life when the rapid

decline in respiratory rate prevents the use of cutoff

values to define normality.

APPENDIX: CENTILE SMOOTHING AND

ESTIMATED EQUATIONS

To construct usable centile curves, the observed raw centiles must be smoothed, fitting interrelated polynomials to them.

Poly-nomials are flexible enough to fit each centile adequately. In our case, a second-degree function was sufficient after log transforma-tion of the data.

Thus, for the ith centile, these polynomials have the following formula:

Yi =A #{247}A,1x + Ax2 (1)

where x is the age in months, and Y Log,0(RR).

This ensures that each centile varies smoothly with age. In order to use all the available information the method constrains the centiles to be interrelated in a smooth way. This is accomplished by modeling the coefficients A11 as polynomial functions of Z,, the

normal equivalent deviate corresponding to the ith centile (for instance, for the 50th centile, Z = .0; for the 95th centile, Z95 = 1.645, etc) according to the following formula:

A11= B0 + B11Z + B12Z12 + .. .+ B1Z. (2)

Combining Equations I and 2 gives the final model.

In our case, for both A. and A1 a first-degree function, and for A2 an intercept-only function were selected. Thus, for the ith centile, omitting for clarity the ith subscript, the equation is:

Log,o(RR) = (Bm + B0,Z) + (B,0 + B11Z)x + B2x2.

For the actual coefficient estimated,

Log(RR) = (1.6801 + 0.08592Z) - (0.01189 + 0.0004301Z)x + 0.0001340x2

for awake data, and

Log(RR) = (1.5773 + 0.09196Z) - (0.01314+ 0.0005155Z)x

+ 0.0001765x2

ACKNOWLEDGMENTS for asleep data.

We are grateful to the medical staff of the Department of Pediatrics of the Ospedale Buzzi, Milano, and to Dr. Aurelio Nova,

Monza, for permission to examine infants and children under their care, and to Alix Green for help in revising the manuscript.

REFERENCES

1. Shann F, Hart K, Thomas D. Acute lower respiratory tract infections in children: possible criteria for selection of patients for antibiotic therapy and hospital admission. Bull WHO. 1984;62:749-753

2. Campbell H, Byass P, Greenwood BM. Simple clinical signs for diag-nosis of acute lower respiratory infections. Lancet. 1988;2:742-743

3. Cherian T, John TJ, Simoes E, Steinhoff MC, John M. Evaluation of

(6)

4. Harari M, Shann F, Spooner V, Meisner S, Carney M, De Campo J.

Clinical signs of pneumonia in children. Lancet. 1991258:928-930 5. Ihff A, Lee VA. Pulse rate, respiratory rate, and body temperature of

children between two months and eighteen years of age. Child Den.

1952;4:237-245

6. Weil JV. Ventilatory control at high altitude. In: Cherniak NS, Widdicombe JG, eds. Handbook ofPhysiology, Sect 3, Vol II. Bethesda, MD: American Physiological Society, 1986:703-727

7. Berman S, Simoes EA, Lanata C. Respiratory rate and pneumonia in

infancy. Arch Dis Child. 1991;66:81-84

8. Richards JM, Alexander JR. Shinebourne PA, de Swiet M, Wilson AJ,

Southall DP. Sequential 22-hour proffles of breathing patterns heart rate in 110 full-term infants during their first 6 months of life. Pediatrics

1984;74:763-777

9. Morley CJ, Thornton AJ, Fowler MA, Cole 1J, Hewson PH. Respiratory rate and severity of illness in babies under 6 months old. Arch Dis Child.

1990;65:834-837

10. Bland JM, Altman DG. Statistical methods for assessing agreement

between two methods of clinical measurement. Lancet. 1986;i:307-310 11. British Standard Institution. Precision of Test MethOds I: Guide for the

Determination of Repeatability and Reproducibility for a Standard Test

Method (BS5497, part I). London: BSI; 1979

12. Emerson JD, Stoto MA. Transforming data. In:Hoaglin DC, Mosteller F, Tukey JW (eds). Understanding Robust and Exploratory Data Analysis.

New York: Wiley & Sons; 1983:97-128

13. Healy MJR, Rasbash J, Yang M. Distribution-free estimation of age-related centiles. Ann Hum Biol. 1988;15:17-22

14. Rasbash J.The GROSTAT II Manual. London: institute of Education,

World Health Organisation; 1989

15. Hoppenbrouwers T, Harper RM, Hodgman JE, Sterman MB, McGinty

DJ. Polygraphic studies of normal infants during the first six months of

life. II.Respiratory rate and variability as a function of state. Pediat Res.

1978;12:120-125

16. Simoes EAF, Roark R, Berman 5, Esler LL, Murphy J. Respiratory rate:

measurement of variability over time and accuracy at different counting

periods. Arch Dis Child. 199166:1199-1203

17. Curzi-Dascalova L, Gaudebout C, Dreyfus-Brisac C. Respiratory fre-quencies of sleeping infants during the first months of life: correlations between values in different sleep states. Early Hum Den. 1981;5:39-54

18. Morley CJ. Respiratory rate and pneumonia in infancy. Arch Dis Child.

1991;66:1001

19. Marchal F, Crance J-P. Measurement of ventilatory system compliance in infants and young children. Respir Physiol. 1987;68:311-318

20. Bryan AC, Wohi MEB. Respiratory mechanics in children. In: Macklem

PT, Mead 1’ eds. Handbook of Physiology, Sect 3, Vol ifi. Bethesda, MD:

American Physiological Society; 1986:179-191

21. Hershenson MB, Cohn AA, Wohi MEB, Stark AR. Changes in the contribution of the rib cage to tidal breathing during infancy. Am Rev Respir Dis. 1990;141:922-925

DOCTORS CUSTOMERS!

Dr. Stephen Jones of Rockville, MD, calls me his

patient.

I like that word; it make

me feel secure. Dr. Richard Seizer, the surgeon and writer, once told a Mayo

Medical School graduating class that the word

patient

comes from the Latin pati, “to

suffer,” adding: “Doctors have patients. This is, above all, what distinguishes us

from lawyers, who have clients . . . We have patients, and they suffer.”

Clients suffer, too, at the hands of some lawyers, but the distinction is valid. In

recent years, however, a dehumanizing note has crept into the medical language:

patients have become

health

care consumers. Victor Cohen, the former Washington

Post health columnist, was among the first to deride the trend toward calling

doctors caregivers and

healthcare

[one word] producers; the new terms lump the MDs

among less well-trained professionals.

The big word now is provider, which has taken care of caregiver.

Thanks to the info explosion, it’s spreading. I used to be a writer. Now I’m a

content provider. Don’t laugh; it could happen to you.

Safine W. The New York Times Magazine. March 1994.

Noted by J.F.L., MD

(7)

1994;94;350

Pediatrics

Antonio Pellegatta, Sultana Razon and Mario Braga

Franca Rusconi, Manuela Castagneto, Norberto Porta, Luigi Gagliardi, Gualtiero Leo,