Socio-demographic characteristics of the study participants
A total of 701 children charts were reviewed and analyzed in this study. Among these more than half of participants (56.49%) were males and 375 (53.5%) of patients were rural resident. The median age of the study participants was 10 months with IQR of 5-24 months and 52.2 % were under the age group of 2-11 months (table 1).
Table 1:- Socio-Demographic Characteristics among children age of 2 up to 59 months admitted with severe community acquired Pneumonia in Public hospitals of Central and North Gondar Zone, Ethiopia January, 2018-December, 2020.
Variables
|
Catagories
|
Frequency (701)
|
Percent (%)
|
Age
|
2-11months
|
366
|
52.21
|
12-59 months
|
335
|
47.79
|
Sex
|
Male
|
396
|
56.49
|
Female
|
305
|
43.51
|
Recedence
|
Urban
|
326
|
46.50
|
Rural
|
375
|
53.50
|
5.2. Nutritional, breast feeding and vaccination status
Of the total 701 study participants, the overall under nutrition was 59(8.42%). Forty-nine (6.99%), 46(6.56%), and 34(4.85%) of them were underweight, stunting, and wasting respectively. About 656(93.58%) children were on vaccination and from these, 317(45.22%) were fully vaccinated. Regarding to breast feeding status from the total study participants 680 (97%), 18 (2.57%) and 3(0.43%) were exclusively breast feeding, had mixed feeding and not breast feed during the first six month of their age respectively (Table 2).
Table 2:- Nutritional, breast feeding and vaccination status among children age of 2 up to 59 months admitted with severe community acquired pneumonia in Public Hospitals of Central and North Gondar Zone, Ethiopia January,2018- December 2020.
Variables
|
Frequency (701)
|
Percent (%)
|
Weight for age
Normal
Underweight
|
652
49
|
93.01
6.99
|
Height for age
Normal
Stunting
|
655
46
|
93.44
6.56
|
Weight for height
Normal
Wasting
|
667
34
|
95.15
4.85
|
Breast feeding status
Exclusive feeding
Mixed feeding
Not breast feeding
|
680
18
3
|
97.00
2.57
0.43
|
Vaccination status
Fully vaccinated
Partially vaccinated
Not vaccinated
|
317
339
45
|
45.22
48.36
6.42
|
5.3. Clinical characteristics, co-morbidities and time to seeking care
Of the 701 study participants 455 (64.9%) were febrile and about 213 (30.39) were hypoxic. The median temperature and oxygen saturation was 37.8 with IQR (37.1-38.6) and 86% IQR (80%-90%) respectively. About 198(28.25%) developed danger signs, and the most frequent danger sign was vomiting 123 (17.55%) followed by grunting 84(11.98%). From the total analyzed children 240(34.24%) had co-morbid and the most common co morbidity, was hyperactive airway disease (childhood asthma) followed by acute gastroenteritis. As regards to the treatment options for the SCAP, 62.91%, 33.38% and 3.71% were treated with ceftriaxone, crystalline penicillin and ampicillin & gentamicin respectively (Table 3).
Table 3:- Clinical Characteristics and co-morbidities among Children age of 2 up to 59 Months admitted with SCAP in Public Hospitals of Central and North Gondar Zone, Ethiopia January,2018- December2020.
Variables
|
Categories
|
Frequency (701)
|
Percent (%)
|
|
|
|
|
Temperature in 0c
|
<37.50c
|
246
|
35.09
|
≥37.50c
.
|
455
|
64.91
|
Oxygen saturation (Spo2)
|
< 90
|
213
|
30.39
|
≥90
|
488
|
69.61
|
Time to seeking care
|
≤ 5 days (early presenters)
|
633
|
90.30
|
>5 days (late presenters)
|
68
|
9.70
|
previous Hx of ARTI
|
Yes
|
133
|
18.97
|
No
|
568
|
81.03
|
Presence danger sign
|
Yes
|
198
|
28.25
|
No
|
503
|
71.75
|
Types of danger signs
|
Vomit every thing
|
123
|
17.55
|
Wheezing
|
56
|
7.99
|
Grunting
|
84
|
11.98
|
Unable to feeding
|
20
|
2.85
|
Lethargic
|
12
|
1.71
|
Loss of consciousness
|
3
|
0.43
|
Types of antibiotics
ceftriaxone
|
441
|
62.91
|
Crystalline penicillin
|
234
|
33.38
|
ampicillin and gentamicin
|
26
|
3.71
|
Presence of Co- morbidity
|
Yes
|
240
|
34.24
|
No
|
461
|
65.76
|
Types of co-morbidity
|
HAAD/bronchial asthma
|
144
|
20.54
|
Acute gastroenteritis
|
66
|
9.42
|
Heart failure /CVD
|
13
|
1.85
|
Anemia
|
5
|
0.71
|
under nutrition
|
21
|
3
|
Pertussis
|
6
|
0.86
|
Measles
|
3
|
0.43
|
Meningitis
|
2
|
0.29
|
HIV/ADIS
|
1
|
0.14
|
Rickets
|
5
|
0.71
|
Others*
|
16
|
2.28
|
Others* renal disease, other acute febrile disease
5.4 The median recovery time and comparisons of survival status between categories of predictors using Kaplan Meier curve and log-rank test.
Of 701 study participants admitted with SCAP, about 688 (98.15%) were recovered. The patients were followed for a minimum of 1 and a maximum of 17 days. The overall median recovery time of children 2-59 age months admitted with severe community acquired pneumonia in public hospitals of central and north Gondar zones was 3 days IQR of 2-5 days. The total person-time risk was 2567 with the overall recovery rate of 26.71 per 100 person days (95% CI: 24.79-28.78) observation.
The overall survival status of children with SCAP was estimated by the Kaplan-Meier survival curve (fig 1)
The Kaplan Meier curve showed that there was understandable difference between groups of predictors. Based on the estimation, patients who had co morbidity, dangerous signs at admission, previous history of ARTI, those children present lately to hospital, those children have rural resident, children had mixed breast-feeding status, and stunted children required long time to recovery from severe community acquired pneumonia compared to their counter parts respectively. For instances, the median time to recovery among children admitted with SCAP who had co-morbidity was 4 days (IQR: 3-5 days) and those children admitted without co-morbidity had the median recovery of 3 days (IQR: 2-4 days). Similarly the median time to recovery of children who had danger sign at admission was 4 days (IQR: 3-5 days), while patients without danger sign was 3 days IQR (2-4 days). Regarding to nutritional status of the child, stunted children had the median to recovery of 5 days IQR (4-7days) while normal children had time to recovery of 3 days IQR (2-4 days) (table 4).
Table 4: Table shows Comparison of survival time to recovery of children with SCAP patients regarding to different categories of predictors by using log-rank test
Variables
|
median recovery time in day with IQR
|
Chi-square
|
Degree of freedom
|
P- Value
|
Residency
|
|
|
|
|
Urban
|
3(2-4)
|
|
|
|
Rural
|
3(2-5)
|
12.77
|
1
|
<0.001
|
Hospital type
|
|
|
|
|
Tertiary hospital
|
3 (2-5)
|
|
|
|
General hospital
|
3(2-5)
|
|
|
|
Primary hospital
|
3(2-4)
|
15.66
|
2
|
<0.001
|
Breast feeding status
|
|
|
|
|
EBF
|
3 (2-4)
|
|
|
|
Mixed feeding
|
5 (3-5)
|
|
|
|
NBF*
|
6 (4-7)
|
15.15
|
2
|
<0.001
|
Duration of seeking care
|
|
|
|
|
Early presenters
|
3(2-4)
|
|
|
|
Late presenters
|
4(3-6)
|
23.74
|
1
|
<0.001
|
Previous hx* of ARTI
|
|
|
|
|
No
|
3 (2-4)
|
|
|
|
yes
|
4(3-5)
|
34.07
|
1
|
<0.001
|
Presence of danger sign at admission
|
|
|
|
|
No
|
3(2-4)
|
|
|
|
Yes
|
4(3-5)
|
34.07
|
1
|
<0.001
|
Presence of co- morbidity
|
|
|
|
|
No
|
3(2-4)
|
|
|
|
Yes
|
4(3-5)
|
52.19
|
1
|
<0.001
|
Types of antibiotics
|
|
|
|
|
Ceftriaxone
|
3 (2-4)
|
|
|
|
Crystalline penicillin
Ampicillin &gentamicin
|
3(2-4)
|
|
|
|
5(4-6)
|
8.91
|
2
|
0.012
|
WFA*
Normal
Under weight
|
3 (2-4)
5(4-7)
|
37.21
|
1
|
<0.001
|
HFA*
Normal
Stunting
|
3(2-4)
5 (4-7)
|
38.56
|
1
|
<0.001
|
WFH*
Normal
Wasting
|
3(2-4)
5 (4-7)
|
17.64
|
1
|
<0.001
|
* hx=history * NBF= not breast feeding *WFA=weight for age *HFA=height for age *WFH=weight for height
There was a significant recovery time difference among categories of residence, Co morbidity, presence of danger sign at admission, antibiotics (drug) type, Hospital type, breast feeding status, duration of seeking care and nutritional status of the child (Weight/Age, Height/Age and Height/Weight) ( Fig 2) .
5.5 model selection and PH-assumption test
Cox regression model fitness and proportional hazard assumptions test were checked graphically and by using Shenfield residual global test. PH assumption was fitted showed global test of (χ2= 21.11, p= 0.0989)
5.6 Predictors of recovery time from Sever community acquired pneumonia
Predictors that have p-value of less than 0.25 in the bi-variable Cox regression analysis were selected and entered in to multivariable Cox regression analysis to identify independent association with time to recovery. Eight predictors (residence, hospital type, duration of seeking care, types of antibiotics given, danger sign at admission, co-morbidity, breast feeding status and height for age (HFA) were statistically significant predictors of recovery time.
According to the multivariable Cox regression analysis, the recovery time was delayed by 20% ( AHR=0.8; 95% CI (0.68-0.94)) among patients of rural residency as compared with its counterpart. Similarly, the duration of recovery time of patients with co-morbidity was increased by 34% (AHR 0.66; 95%CI (0.56-0.79)) as compared with their counter parts. The hazard of children who had danger sign at admission for delay recovery time was 19% (AHR; 0.81, 95% CI (0.678-f 0.66; 95%CI (0.56-0.79)) times as compared with those admitted without danger sign. Regarding the duration of seeking care children with SCAP who came lately (≥ 5 days) to the health facility were 29% (AHR= 0.71:95% CI (0.541-0.928)) delayed recovery time as compared to those who came early (< 5 days).
Those children who were on mixed feeding and not breast feeding was 47% (AHR=0.53; 95% CI (0.32-0.87)) slow their recovery time as compared to exclusive breast feed children. Likewise, the recovery time of stunted children from SCAP decreased by 37% as compared to children with normal height for age (AHR; 0.63, 95% CI (0.44–0.89)).
Regarding of the treatment regimen, the recovery time of children treated with ampicillin and gentamicin for SCAP was decreased by 39% as compared to children treated with ceftriaxone (AHR, 0.61, (95% CI (0.40–0. 92)). The recovery time of children admitted with SCAP at primary hospital was 1.43 times shorter (AHR=1.43; 95% CI (1.16-1.76) than that of children admitted at specialized hospital (Table 5).
Table 5:- Multivariable Cox Proportional Regression Analysis For Independent Predictors Of Recovery Among Children Age 2-59 Months With SCAP Admitted In Public Hospitals Of Central And North Gondar Zone, Ethiopia From January,2018 to December,2020
Variables
|
Categories
|
CHR(95% CI)
|
AHR(95% CI)
|
Residency
|
Urban
|
1
|
1
|
Rural
|
0.80(0.69-0.93)*
|
0.80(0.68-0.94)*
|
Hospital type
|
Tertiary
|
1
|
1
|
General
|
1.03(0.87-1.23)
|
0.96(0.80-1.15)
|
Primary
|
1.36(1.11-1.64)*
|
1.43(1.16-1.76)**
|
Duration of care
|
< 5 days
|
1
|
1
|
≥5 days
|
0.599(0.46-0.78) **
|
0.709(0.541-0.928)*
|
Breast feeding status
|
EBF
|
1
|
1
|
Mixed
|
0.52(0.32-0.85)
|
0.53(0.32-0.87)*
|
NBF
|
0.41(0.13-1.27)
|
0.37(0.11-1.16)
|
Previous hx of ARTI
|
No
|
1
|
1
|
Yes
|
0.87(0.71-105)
|
0.91(0.74-1.11)
|
Danger sign
|
No
|
1
|
1
|
Yes
|
0.667(0.56-0.79)**
|
0.81(0.68-0.97)*
|
Co morbidity
|
No
|
1
|
1
|
Yes
|
0.61(0.52-0.72)**
|
0.66(0.56-0.79)**
|
Types of antibiotics given
|
Ceftriaxone
|
1
|
1
|
Crystalline penicillin
|
0.99(0.85-1.17)
|
0.98(0.83-1.15)
|
Ampicillin &gentamicn
|
0.61(0.41-0.92)*
|
0.61(0.40-0.92)*
|
Weight/age (WFA)
|
Normal
|
1
|
1
|
Underweight
|
0.48(0.35-0.65)**
|
0.69(0.47-1.00)
|
Height/age (HFA)
|
Normal
|
1
|
1
|
Stunted
|
0.46(0.34-0.63)**
|
0.63(0.44-0.89)*
|
Weight/height (WFH)
|
Normal
|
1
|
1
|
Wasted
|
0.55(0.39-0.78)**
|
0.75(0.49-1.11)
|
NBF-=not breast feeding, EBF =Exclusive breast feeding, hx =history
* p- Value < 0.05 ** p –value ≤ 0.001