Perinatal Epidemiology Workshop

Russell S. Kirby, PhD, MS, FACE

University of Wisconsin Medical School

As of July 2002: Department of Maternal and Child Health School of Public Health University of Alabama-Birmingham

Objectives
Focus on components of infant mortality Review the theoretical basis for conventionally used measures (neonatal, postneonatal) Evaluate the utility of these measures with populationbased data Consider alternative methodologies for operationalizing these measures Propose study designs to test alternative measures in relation to conventional measures

TOP TEN LIST

TEN BEST WAYS TO MISUSE CONFIDENCE INTERVALS
With apologies to David Letterman, and thanks for editorial assistance to Elizabeth Kirby and for their insights to the following Internet contributors:
Patrick Remington, University of Wisconsin Medical School Robert Meyer, N.C. Birth Defects Monitoring Program, State Center for Health Statistics N.C. Division of Public Health Richard Miller, Wisconsin Bureau of Health Information Russel Rickard, Colorado Responds to Children with Special Health Needs Kim Hauser, University of South Florida

R.S. Kirby, March 2002

"If the confidence interval is very tight, the case for causation is strengthened...."
Submitted without attribution as a quotation from a manuscript under review by Patrick Remington University of Wisconsin Medical School February 24, 2002

Top Ten List: Ten Best Ways to Misuse Confidence Intervals

Number 10 Say It With Total Confidence

Being a statistician means never having to say you're wrong. Don't be afraid to use those 100% CIs so that you can state with authority:

"I'm 100% sure that the true population parameter lies between zero and infinity!"

Top Ten List: Ten Best Ways to Misuse Confidence Intervals

Number 9 Out of Sight, Out of Mind

Whether you calculated the CIs or not, it isnt necessary or desirable to include them in your publication.
A general statement in the text to the effect that all statistical values were significant should be sufficient for all but lay audiences.

Top Ten List: Ten Best Ways to Misuse Confidence Intervals

Number 8 A Picture is Worth a Thousand Words

Assess statistical significance through visual comparison of confidence intervals. For example: which of the following confidence intervals is larger? <-------> OR >-------<

Top Ten List: Ten Best Ways to Misuse Confidence Intervals

Number 7 Smoke and Mirrors

On a graph, overly wide confidence bands can be adjusted by plotting the data on an arithmetic scale and the CIs on a logarithmic scale. Your friends will be amazed at how tight those CIs become.

Top Ten List: Ten Best Ways to Misuse Confidence Intervals

Number 6 How Big is Your N?

Remember that any results would have been statistically significant if only your sample size was large enough. This should not ruin otherwise good science.

To compensate for small sample sizes, adjust your CIs so that the null value is always excluded. Don't be afraid to use 15% CIs if necessary, if that will help support your hypothesis.

Top Ten List: Ten Best Ways to Misuse Confidence Intervals

Number 5
To Really Foul Things Up . . .
Never include your point estimate within the confidence limits. When questioned, blame it on the computer program.

Top Ten List: Ten Best Ways to Misuse Confidence Intervals

Number 4 Will 99 and a Half Do?

95% CIs are trite and commonplace. Be creative Try reporting results such as the 2 log(67.45%) CIs and see what exciting results you get.

No one will understand it, but they dont understand the 95 or 99% CIs either.

Top Ten List: Ten Best Ways to Misuse Confidence Intervals

Number 3 Sound Bytes Are Best

Selectively quote the results concerning one confidence limit. For example, when the estimate of the confidence interval is 0 to 70, perhaps for the association between watching late night reruns and sleep disorders: This risk factor decreases the risk of the outcome up to 70%.

Top Ten List: Ten Best Ways to Misuse Confidence Intervals

Number 2 Why Infer When You Can Guess?

To demonstrate statistical significance for a comparison of rates or proportions, visually examine the two confidence intervals.

Try comparing the lower confidence bound of the smaller value with the upper confidence bound of the larger value. If that still doesnt work, try fitting ever narrowing confidence bands (e.g. the 15% CIs as in Number 6 above) until the confidence limits no longer overlap.

Top Ten List: Ten Best Ways to Misuse Confidence Intervals

Number 1 Is That All, Folks?

The best way to do misuse confidence intervals is to confuse statistical significance with true, substantive significance. An observed statistical difference begs the question So what? which is too seldom asked.

Top Ten List: Ten Best Ways to Misuse Confidence Intervals

If you pay strict attention to these suggested methods for using confidence intervals, you too can misuse confidence intervals with confidence. -- R. S. Kirby March 25, 2002

Brief Summary for Those Who Are Knitting, Doing Crossword Puzzles, or Discerning the Geometric Pattern in the Carpeting
Too often in public health, we use measures and present data in familiar forms because thats the way weve always done it. The concepts underlying measures of neonatal and postneonatal mortality have their origins in sociological theory from the period 1880-1940. Periodically our indicators and outcome measures need to be reviewed for contemporary validity (especially construct validity). Some alternatives (for infant mortality classifications) include more careful consideration of underlying causes of death, more sophisticated record linkages, and developmental measures. This is fertile ground for innovative, multidisciplinary research.

Resource
Kirby, Russell S., "Neonatal and Postneonatal Mortality: Useful Concepts or Outdated Constructs?", Journal of Perinatology, 13,6 (November-December 1993), 433-441.

Historical Context
Neonatal and postneonatal mortality developed as as demographic measures Proxies for general categories of cause of infant death
Endogenous causes (perinatal, congenital and immediate newborn period) Exogenous causes (nonperinatal infection, injury, homicide, other external causes)

Conventional Wisdom
Deaths occurring early in the first year of life are more likely associated with endogenous causes Deaths after the early part of infancy are more likely associated with endogenous causes (socioeconomic or standard of living) Since the 1940s, we have continued to define neonatal mortality as deaths in the first 27 days of life, and postneonatal mortality as deaths in the balance of the first year of life

Whats Happened Since 1940

Secular decline in infant mortality rates Since the 1970s, a more rapid decline in neonatal mortality rates
Prior to about 1985, much of this decline resulted in postponement of death into the postneonatal period More recently, technological advances and improvements in neonatal resuscitation and NICU management have led to greater relative declines in neonatal mortality In the 1990s, some health promotion programs have been successful in targeting SIDS, a condition that occurs primarily in the postneonatal period

Neonatal and Postneonatal Mortality Rates United States, 1960 - 1999

Rate per 1,000 live births

30 25 20 15 10 5 0

26.0

24.7 20.0 16.1 12.6 10.6 9.2

7.6

7.1

1960

1965

1970 Infant

1975

1980

1985

1990

1995

1999

Neonatal

Postneonatal

Source: National Center for Health Statistics, final mortality data Prepared by March of Dimes Perinatal Data Center, 2002

Methods for the Useful Constructs Study

Information on deaths from linked birth and infant death certificates Underlying causes of death classified into endogenous and exogenous causes Mortality rates compared within birth weight strata between endogenous and neonatal and between exogenous and postneonatal mortality rates

The Bottom Line

At least in the 1980s, neonatal and endogenous mortality measures are fairly synonymous Postneonatal mortality is no longer a useful proxy for exogenous infant mortality This renders the conventional classification suspect for current purposes New on the scene are advances in perinatal-infant medicine and nursing management, the new ICD-10 classification for diseases, and more readily available data resources through record linkage, chart review, and other data acquisition methods

Future Studies?

Three Alternative Formulations: One

Use conventional definitions, but calculate neonatal and postneonatal mortality rates bases on corrected gestational age
Thus, a term neonate who dies during the first month after birth is a neonatal death A 32 week infant who dies prior to the 12th week of life (40-32=8 + 4 weeks of conventional neonatal period) is considered a neonatal death

Three Alternative Formulations: Two

Include fetal deaths in the analysis, regarding all of these events as non-postneonatal. Conduct the analysis conventionally, or use the corrected gestational age definition

Three Alternative Formulations: Three

Incorporate linked birth-infant death certificate and infant hospital discharge records into the analysis Define neonatal and postneonatal in relation to initial continuous hospital stay.
Example: a baby born at one hospital who is transferred to an NICU at another hospital and is discharged at 9 weeks has a neonatal period lasting 63 days. Another example: a healthy newborn is discharged on the second postpartum day, and has a neonatal period lasting 2 days

Three Alternative Formulations: Three (continued)

Use the person-days method for calculating rates. Thus, the neonatal mortality rate is the number of neonatal deaths according to this definition, divided by the total number of initial newborn stay days. The postneonatal mortality rate is the number of non-neonatal infant deaths, divided by the balance of lived postneonatal days.

However beautiful the strategy, you should occasionally look at the results.
-- Sir Winston Churchill

Critique of Alternatives

Critique of Alternatives
Corrected gestational age will only work when gestation is measured accurately. That is a topic for at least another workshop if not an entire MCH epidemiology conference but allowing for some error at early gestational ages its worth a try All of the problems with birth certificate and infant death certificate data quality are at least an order of magnitude worse with fetal death records

Critique of Alternatives
Hospital discharge databases are becoming more widely available. However, in most states these files are not linked in the manner necessary to support the proposed analysis. Most public health professionals and demographers are unfamiliar with rates expressed as person-time measures A validated measure for neonatal and postneonatal mortality could serve as the basis for more careful multivariate analyses, but first its validity and reliability must be independently verified For the time being, it is likely we are stuck with the same old same old.

Questions?
Im interested in hearing from you! I can be reached by e-mail at r-kirby@whin.net 414-219-5610, FAX 414-219-5201 Ill provide forwarding information at these locations after July 2002