Study selection and data extraction
The search strategy identified 80 articles from PubMed, 60 articles from Google Scholar, 45 articles from Cochrane Library, 10 articles from African Journals Online, 7 articles from Ethiopian’s University online library and 5 articles by manual search. Of which, 134 were excluded due to duplication, 35 through review of titles and abstracts. Additionally, 31 full-text articles were excluded for not reporting the outcome variable and other reasons. Finally, 19 were included to the prevalence and/ or associated factor analysis on partograph utilization [Fig.1].
Study characteristics
Different factors such as; midwifery profession, female health care providers, Emergency Obstetric and Newborn Care training, Knowledge of partograph, short term training on partograph, favorable attitude towards partograph, and working at the health center were included in this study. Nineteen cross-sectional studies with a total of 6237 obstetric care providers were included in this review. All of them were conducted at facility-based by study setting. Regarding the study area, six of the studies were conducted at SNNPR (south nation nationalities and people representative), four in Tigray, Amhara, Oromia; each account three studies respectively (Table 1).
Table 1: Descriptive summary of fifteen included studies in the systematic review and meta-analysis
Author (year of study) (reference number)
|
Sample size
|
Response rate
|
Study region
|
P(95% :CI)
|
Quality of score
|
Fantu A.et al(2012)[15]
|
381
|
88
|
Amhara
|
29(24,33)
|
8
|
Habtamu R.et al(2017)[16]
|
224
|
90.2
|
Oromia
|
89(85,93)
|
9
|
Wakeshe W.et al(2015)[17]
|
266
|
97.4
|
Oromia
|
84(80,88)
|
9
|
Negash W.et al(2013)[18]
|
403
|
94.5
|
Amhara
|
40(35,45)
|
8
|
Haymanot M. et al(2015)[13]
|
441
|
98
|
Dire Dawa
|
92.6(90,95)
|
8
|
Tesfay H. et al(2017)[19]
|
220
|
90
|
Tigray
|
73(67,79)
|
8
|
Desalegne A. et al(2015)[20]
|
273
|
100
|
Amhara
|
53(48,60)
|
9
|
Kidist E. et al(2016)[21]
|
300
|
93.3
|
SNNP
|
51(45,57)
|
9
|
Kidest G.et al(2016)[22]
|
442
|
99
|
SNNP
|
73(68,78)
|
9
|
D.Markos et al(2014)[23]
|
401
|
91
|
SNNP
|
70(66,75)
|
8
|
Engida Y.et al(2012)[24]
|
202
|
96.5
|
Addis Ababa
|
57(50,64)
|
9
|
Sena B.et al(2012)[14]
|
340
|
80.6
|
Oromia
|
6.9(4,10)
|
9
|
Gutema C. et al(2015)
|
309
|
89
|
SNNP
|
54(48,59)
|
9
|
Daniel B.et al(2016)[25]
|
127
|
100
|
SNNP
|
26(18,34)
|
9
|
Haftom G et al(2015)[26]
|
233
|
93
|
Tigray
|
57(51,64)
|
8
|
Guesh W.et al(2018)[27]
|
414
|
98.1
|
Tigray
|
83(31,43)
|
8
|
Yosef Haile.et al(2019)[28]
|
436
|
95
|
SNNP
|
55.4(2,9)
|
9
|
Tesfay H. et al(2019)[19]
|
220
|
98
|
Tigray
|
73.3(21,32)
|
9
|
Azeb A.et al(2017)[29]
|
605
|
98.1
|
Addis Ababa
|
69(36,45)
|
9
|
Utilization of partograph in Ethiopia
In resource-limited countries like Ethiopia Partograph has been promoted by the World Health Organization as the "gold" standard for assessing the progress of labour, and as a decision-making tool. The overall pooled prevalence of partograph utilization among obstetric care providers in Ethiopia was presented with a forest plot (Fig. 2). The pooled estimated prevalence of partograph utilization among obstetric care providers in Ethiopia was 59.95% (95% CI:46.8,73,I2 = 99.4%, P < 0.001)
Publication bias
Funnel plot was assessed for asymmetry distribution of prevalence of partograph utilization among obstetric care providers by visual inspection (Fig. 3). Egger’s regression test showed with a p-value of 0.02 indicated the existence of publication bias.
Sensitivity analysis
This systematic review and meta-analysis showed that the point estimate of its omitted analysis lies within the confidence interval of the combined analysis. Therefore, trim and fill Analysis was no further computed (fig. 3).
Subgroup analysis
Subgroup analysis was employed with the evidence of heterogeneity. Furthermore, the Cochrane I2 statistic =99.4%, P < 0.001) with evidence of marked heterogeneity, subgroup analysis was done by using year of study and study area (Table 2).
Table 2: Sub-group analysis of the prevalence of partograph utilization in Ethiopia by region and year of publication.
Subgroup
|
No. of studies
|
Model
|
Prevalence (95% CI)
|
I2 (%)
|
p-value
|
By region
|
Amhara
|
3
|
random
|
40.89 (27.40, 54.38)
|
95.3
|
<0.001
|
Oromia
|
3
|
random
|
60.1(1.75, 118)
|
99.9
|
<0.001
|
Addis Ababa+Dire Dawa
|
3
|
random
|
73(52.56,93.7)
|
98.8
|
<0.001
|
Tigiray
|
4
|
random
|
71.9 (61,82.59)
|
93.9
|
<0.001
|
SNNP
|
6
|
random
|
55.10 (43.6, 66)
|
97.4
|
<0.001
|
By publication year
|
|
2012-2014
|
5
|
random
|
40.67 (15.61, 65.72)
|
99.4
|
<0.001
|
2015-2017
|
11
|
random
|
65.93 (55.06, 76.81)
|
98.4
|
<0.001
|
2018-2019
|
3
|
random
|
70.59 (53.28, 87.91)
|
97.6
|
<0.001
|
Associated factors of partograph utilization
A total of 19 studies were included for analysis of associated factors of partograph utilization. We identified seven main associated factors with the pooled odds ratio ranging from 2.12 to 5.66. These associated factors were midwifery profession, female health care provides, Emergency Obstetric and Newborn Care training, attitude, Knowledge of partograph, health center, and short term training on partograph, favorable attitude towards partograph, and working at health center.
Midwifery profession on partograph utilization
The result of this systematic review showed a significant association between midwifery profession and partograph utilization. Obstetric care providers who were midwifery in profession were 4 times more likely to utilize partograph as compared to other Obstetric care providers (AOR: 3.97,95%CI:2.6,6, p=0.19, I2 =28.8%)(Fig.4).
Emergency Obstetric and Newborn Care training
This study also revealed that there is a significant association between Emergency Obstetric and Newborn Care training and partograph utilization. Those obstetric care providers who received Emergency Obstetric and Newborn Care training were around 3 times more likely to utilize partograph as a decision-making tool (OR = 2.9,95% CI:2.19, 3.83) (Fig. 6).
Training on partograph and partograph utilization.
Having short term training on partograph showed statistically significant association with partograph utilization among obstetric care providers.
Obstetric care providers who had training on partograph were around 6 times more likely to utilize partograph (O=5.7,95%CI:2.5,12.9) as a decision-making tool as compared to those obstetric care providers who did not receive training on partograph(Fig. 7).
Attitude of obstetric care providers on partograph utilization
Obstetric care providers who had favorable attitude towards partograph utilization were more likely to utilize partograph as a decion making tool (AOR=2.3,95%CI:1.5,3) as compared to obstetric care providers who had unfavorable attitude towards partograph(Fig. 9).
Working at health center on partograph utilization
Additionally, working at health center was significantly associated with partograph utilization among obstetric care providers. Obstetric care providers who worked at health center were around 4 times(AOR=3.5,95%CI:2.5,4.9) more likely to use partograph as compared to obstetric care providers worked in Hospitals (Fig. 10).