Factors associated with zinc prescription practice among children with diarrhea who visited public health facilities in Addis Ababa, Ethiopia: A cross sectional study

According to World Health Organization’s (WHO) diarrhea is de ined as passage of frequent watery or loose stool at least 3 times in 24 hours [1]. The disease is multifactorial and affects considerable number of children. Globally, every year about 1.7 million diarrheal cases and 525, 000 deaths are estimated [2]. Rotavirus has been recognized as a common infectious agent, and it has been addressed through the provision of vaccine to children through the routine immunization program. Though prevention and treatment modalities are there to manage diarrheal cases, it continued signi icantly Abstract


Introduction
According to World Health Organization's (WHO) diarrhea is de ined as passage of frequent watery or loose stool at least 3 times in 24 hours [1]. The disease is multifactorial and affects considerable number of children. Globally, every year about 1.7 million diarrheal cases and 525, 000 deaths are estimated [2]. Rotavirus has been recognized as a common infectious agent, and it has been addressed through the provision of vaccine to children through the routine immunization program. Though prevention and treatment modalities are there to manage diarrheal cases, it continued signi icantly to contribute for under ive children mortality. About 22% of mortality among under-ive children was attributed to diarrhea [3]. Ethiopia is one of the top ive countries accounted for high burden of under-ive diarrheal deaths [4].
Zinc is a micro-mineral with a range of health bene its [5]. It is one of the public health important micronutrients [6]. Animal source foods, sea animal based lesh foods and legume crops are food items that can serve as good sources of zinc [7]. Land locked countries like Ethiopia are not naturally favored for sea foods. In the country the consumption of animal source foods is far lower than developed countries https://www.heighpubs.org/japch 028 https://doi.org/10.29328/journal.japch.1001026 [8]. It is also documented that small proportions of Ethiopian children consumed animal source foods [9][10][11]. Unless household level implementable food processing techniques like soaking, germination and fermentation are well practiced, bioavailability of micronutrients in grains including legumes/ pulse crops was con irmed to be low for that they contain antinutritional factors [12]. But pulse/legume crop processing practices were recommended for further study [13] and they were under promotion [14,15]. Evidence also igured out that complementary foods in Ethiopia were short of providing zinc to the level that growing children require [16].
Illness and nutrition status affect each other [17]. Likewise, diarrhea can happen and get worsen because of nutritional derangements, and it can also lead to nutritional de iciencies [18]. Accordingly, World Health Organization (WHO) and United Nations Children Fund (Unicef) recommended the inclusion of zinc in the treatment of childhood diarrhea [19]. However, evidences are limited on zinc prescription practice, perceived cost and willingness to pay for and related factors. Therefore, the purpose of the current study was to assess zinc prescription practice and identify related factors among children with diarrhea who visited public health institutions in Addis Ababa, Ethiopia.

Study area, design and period
The study was conducted in Addis Ababa city administration, Ethiopia. Based on the 2007 census, the population of Addis Ababa was counted to be 3,384,569 million [20]. Addis Ababa city has a total of 10 sub-cities and 86 health centers. A cross-sectional epidemiologic study design was applied. The study was conducted from September 23-November 20, 2015.

Sampling
Sample size was determined using a single population proportion formula [21] considering 95% con idence level, 50% proportion, 1.5 design effect and 10% non-response rate. Multi-stage sampling technique was employed to reach at the study unit level. Two health centers were selected from each of the ive selected sub-cities. The selection of both sub-cities and health center was on lottery bases. A total of 633 underive children with diarrhea were included in the study who attended the selected health centers.

Data collection
Data were collected using a pre-tested semi-structured questionnaire. The tool consisted of questions on sociodemographic and economic characteristics, and zinc and related issues relevant to the objectives of the investigation. Eight trained data collectors and two supervisors participated in the data collection process. Data were collected through one-on-one interview.

Data management
Each questionnaire was checked for completeness, missing values and unlikely responses and cleaned manually. Questionnaires judged incomplete were excluded from analysis. Data entry was performed using Epi Info version 7 and analyses was performed using SPSS version 20.0 software. During data analysis sorting was used to check for missed variables, outliers and other errors. Descriptive statistical analyses were carried out to generate summary igures on the socio-demographic and demographic characteristics and zinc related outcomes. Bivariate and multivariable logistic regression analyses were done to identify factors associated with zinc prescription. Signi icance of results was decided at p -value of 0.05.

Sociodemographic and economic characteristics
The study was done on a total of 633 under ive children. Twenty four (n = 24) incomplete data cases were excluded from analysis. Slightly more than half, 334 (54.8%), of the children were males. Children's mean age was 19.5 months.

Factors associated with zinc prescription
The multivariable logistic regression analysis showed that children with caretakers who had used zinc before were 2.3 times more likely to receive zinc prescription for diarrhea than their counter parts (AOR = 2.3; 95% CI: 1.34-4.01). Moreover, children whose caregivers had exposure to zinc related information were 2.6 times more likely to receive zinc prescription (AOR = 2.6; 95% CI: 1.53-4.60). This study also revealed that zinc was 6.9 times more likely (AOR = 6.9; 95% CI: 3.84-12.6) to be prescribed for children with diarrhea who were with caregiver willing to pay for zinc (Table 4).

Discussion
This study was conducted among pairs of caretakers and children with diarrhea to assess zinc prescription practice and identify associated factors in Addis Ababa, Ethiopia. Based on the inding of the current study, 62.1% of children received zinc prescription. Previous zinc usage, care givers exposure to zinc related messages, willingness to pay for zinc were variables associated with zinc prescription.
The proportion of children for whom zinc prescribed was 62.1%. The proportion was higher than igures from studies conducted in Nepal (15.4%) [22] and India (22%) [23]. But it was lower than the indings from study done in Kenya (64.30%) [24] and in Mali (66.6%) [25]. This may be due to previous zinc usage, exposure to message about the use of zinc as treatment for diarrhea, willingness to pay for zinc, socio economic variation and the type of health facilities the participants visited.

Awareness about zinc and zinc source and zinc former usage
Pertaining to the exposure to message on zinc, 484 (79.5%) of the respondents heard about zinc. Health professionals were the major (63.6%) source of information about zinc, while radio/television/other posted materials also contributed. More than half (60.1%) of the respondents were aware that they can get zinc from health center, and private pharmacy was also mentioned by more than one fourth (27.6%) of the study participants as place where to get zinc. Zinc was used only by 35% of the study participants previously for diarrheal treatment. Of the 609 respondents, almost three-fourth (73.9%) of caregivers was willing to pay for zinc. There were participants (25.1%) who looked for charge free zinc (cost to pay ranging 0.00 -2.5 $US). The median estimated cost of which participants were willing to pay for zinc was 6ETB [0.30 ±0.45 $US] ( Table 2).

Zinc administration
Zinc was not prescribed for more than one-third (37.9%) of children who attended health institutions because of diarrhea ( Figure 1).
Out of the total zinc prescriptions, majority of zinc prescriptions (75.1%) were written by female health workers and professional nurses (52.9%). Almost one-fourth (24.6%) of the caretakers obtained zinc for free from the health centers   that is serve as treatment for diarrhea. The result is higher than the indings of studies from Nepal (53%) [22], Uganda (17.2%) and Benin (72.1%) [26]. This may be due to the type of exposure to message on zinc treatment for diarrhea from health professionals and mass media.
Two-third (75.1%) of zinc prescription was written by female health care workers. Consistent to the inding of the current study, a research conducted in Nigeria also reported that much of the zinc prescriptions were by female health care workers (38.1%) [29]. With regard to health cadres with whom did the prescription were written, 65.8% were by Nurse, 32.3% by health of icers and 1.9% general practitioners. In Kenya, 62.5% of the prescriptions were written by community health of icers [27]. The difference might be due to the number dominancy of nurses in Ethiopian health care system and difference in health care system structure and cadre organization between the countries.
Children from caretakers exposed to messages about zinc from health care workers were 2.6 times (AOR = 2.6, 95% CI: 1.53-4.60) more likely to receive zinc prescription.
The association is consistent with report of a study done in Kenya in which those who had recent contact with health care workers were reported for their higher chance receiving of zinc prescription. A study conducted in Nepal also strengthens the positive association between exposure to zinc related message and the chance of zinc prescription. It reported 2 times (AOR = 2.02, 95% CI: 1.85-2.19) more chance for those who were exposed to messages from media [22,30].
This study also revealed that children of caretakers who were willing to pay for zinc were seven times (AOR = 6.9; 95% CI: 3.84-12.6) more likely to receive zinc prescription than their counter parts. In agreement to the current inding, a study conducted in Kenya reported importance of caretakers willing for zinc usage as treatment of childhood diarrhea. In the Kenyan study caretakers in the zinc-user group were willing to pay more for a course of zinc in the future than nonuser group [30].

Conclusion
Even though zinc prescription is recommended to be included in the integrated management of childhood illness for the management of diarrhea, in the study area it was not prescribed for considerable proportion of children with diarrhea. The proportion of caretakers who were not willing to pay for zinc and the need for charge free health service including zinc might have contributed for no zinc prescription. Previous experience in use of zinc for diarrheal management, exposure to zinc related message and willingness to pay for zinc were identi ied being positively associated with zinc prescription. Health care workers shall be encouraged to prescribe zinc to all children with diarrhea. Intervention to increase willingness to pay for zinc and zinc bene it communication shall be strengthened in parallel with operational researches

Ethics approval and consent to participate
Ethical clearance was obtained from ethical clearance committee of Debremarkos University. A written letter of requesting for cooperation was also handed to all of the included health centers. Oral consent was secured from each of the study participants. Their participation was on voluntary bases. Obscurity was assured throughout the data collection and management process.

Funding
Cost of data collection was covered by the authors.

Authors' contribution
AW and MB: conception and design of the study; AW, MB and AKD: Data analysis, interpretation of results; AKD: manuscript drafting. AW and MB: Manuscript review. All authors read and approved the inal manuscript.