Table 6.

Final negative binomial regression models for prospective rates of acute chest syndrome^* and vasoocclusive pain^† in children with sickle cell anemia

	IRR	95% CI	P Value
Prospective rates of ACS
Retrospective rate of ACS events/yr	14.14	7.38–27.07	<0.001
Obstruction	0.89	0.42–1.88	0.76
Restriction	1.01	0.48–2.11	0.98
Nonspecific	0.56	0.22–1.40	0.22
Prospective rates of pain
Age, yr	1.07	1.00–1.15	0.041
Retrospective rate of pain events/yr	2.25	1.80–2.80	<0.001
Obstruction	0.70	0.36–1.37	0.30
Restriction	0.59	0.33–1.07	0.08
Nonspecific	0.92	0.42–2.02	0.84

Definition of abbreviations: ACS = acute chest syndrome; CI = confidence interval; IRR = incidence rate ratio.

Median length of follow up, 4.6 years; range 3 months–6.7 years. N = 136 with complete data. See Table E2 in the online supplement for analyses of the 121 participants with ≥12 months of prospective follow up.

^*Initial screening model of the association between lung function pattern and prospective ACS was adjusted for: sex, prior history of ACS, white blood cell count, reticulocyte count, asthma, history of wheezing leading to shortness of breath, bronchodilator responsiveness, and use of inhaled corticosteroids.

^†Initial screening model of the association between lung function pattern and prospective pain was adjusted for: age, sex, prior history of pain episodes, hemoglobin, reticulocyte count, white blood cell count, bronchodilator responsiveness, wheezing leading to shortness of breath, early life environmental tobacco smoke, and ln (IgE).