Data Management Knowledge, Practice and Associated Factors of Ethiopian Health Extension Workers in Gamo Gofa Zone, Southern Ethiopia: A Cross-Sectional study

Public health decision-making is seriously reliant on the timely availability of sound data. The role of health information systems is to generate, analyse and disseminate such data [1,2]. At the level of individuals and communities, information is needed for effective clinical management and for assessing the extent to which services are fulfilling community demands. At the higher level of the health system, health information capacitate health planners and managers to take decision regarding the effective functioning of health facilities and of the health system as a whole [3].

numerical and statistical skills is critical. A study done in North Gondar revealed that lack of skilled and trained manpower with deep knowledge of public health, competency on data collection, analysis and indicator calculation contributes for poor performance of existing health information system (HIS) in Ethiopia [8]. Assessment of utilization of HMIS done in north part of Ethiopia revealed that, out of 84.3% data collected daily, only 22.5% were utilized and 17.7% units of HIV/AIDS changed their data into information at district and facility level and used it for immediate decision making. From the total study units only 13.2% properly document their reports and registration books [9].
While data management knowledge and practice is critical at all level of the health workers in the use of information for decision making, [1] there was no previous studies which investigated the data management knowledge and practice of HEWs and associated factors in the study area. Therefore; this study will have greater input to program managers for designing programs, proper implementation and evaluation of their contribution. It could serve as base line for further study.

Methods
Institution based cross-sectional study was conducted from March 1-20 2012 among HEWs who have been working in Gamo Gofa Zone The study area is, one of the Zones of Southern Nation Nationality and People Regional (SNNPR) state of Ethiopia. The Zone covers 12,003.79 square kilometres ranging from low-arid to high-land (600 m-4550 m) areas. Gamo Gofa Zone is divided into 15 wored as and 2 town administration which are grouped into 448 rural and 34 urban kebeles. The total population of this Zone is estimated to be 1,837,896 most of them living in rural. The three zonal hospitals, 65 health centres, 466 health posts, 832 Rural Health Extension Workers (RHEWs) and 70 Urban Health Extension Workers (UHEWs), Clustering Sampling technique was employed to select the study participants. The sample size required (457) for this study was determined using single population proportion formula by considering the following assumptions: prevalence of data management knowledge 50% (considering with no prior study), 95% level of confidence, 5% margin of error, design effect of 2 and 10% non-response rate. The numbers of clusters needed were selected by using lottery methods. All urban and rural HEWs who were working in the randomly selected wored as (clusters) were included. Eight (BSc in environmental health and health officers) coordinators were trained and assigned for data collection process.
Structured self-administered questionnaire was used to collect data. The questionnaire was prepared in English and it was translated to Amharic language (local language) to make it appropriate and easily understandable for the study participants. The Amharic version was again translated back to English to check the consistency of the meaning. As the HEP is a new program and has been practiced only in Ethiopia we could not find standardized questionnaire from national and international literatures. So we have prepared the questionnaire by looking Health Extension Workers job profile, in consultation with Health Extension Workers, program managers and pre-testing the questionnaire. Most of the questions were closed-ended with pre-coded responses, mainly grouped into socio-demographic characteristics, knowledge of data management and practice of data management, organizational factors and technical factors.

Operational definition
1. Good knowledge: HEWs who scored five points and above out of ten questions prepared for assessment of knowledge.
2. Poor knowledge: HEWs who scored below five points out of ten questions prepared for assessment of knowledge.
3. Good practice: HEWs who scored six points and above out of nine questions prepared for assessment of practice.
4. Poor practice: HEWs who scored below six points out of nine questions prepared for assessment of practice.

Data management knowledge
It focus on the specific knowledge of knowing how to collect, process, analyse data, knowing how and where to report the finding for the sixteen basic health packages collected by Health Extension Workers as their routine day to day work.

Data management practice
Focus on the ability to collect, process data, analyse data, able to present the finding using table, graphs for the sixteen basic health packages collected by Health Extension Workers as their routine day to day work.
Data were entered into Epi-Info Version 3.5.1 and exported to SPSS Version 20 for analysis. Frequencies, proportions and summary statistics were used to describe the study population in relationship to the relevant variables. Bivariate and multivariate logistic regression was computed out to see the effect of each independent variable on the dependent variable. Odds ratio and 95% CI were computed to assess the strength of the association and the statistical significance. A p-value of less than 0.05 was used for statistical significant associations The study was approved by University of Gondar, College of Medicine and Health Science, Institute of Public Health Ethical Committee, and written informed consent was obtained from the study participants.

Results
Four hundred twenty one HEWs participated in the study with a response rate of 92.1%. The mean age of participants was 24.4 ± 3.33 years. Of the total participants, 296 (70.3%) were Protestants and 390 (92.6%) were from the Gamo ethnic group. 258 (61.3%) were married. Most of the participants (92.2%) were rural HEWs. More than fifty percent (56.3%) had work experience of 4-6 years ( Table 1).

Data management knowledge
Of the 421 participants, 245 (58.2%) respondents have good knowledge of data management, 402 (95.5%) respondents reported that they knew data processing, and almost all (98.6%) respondents know to whom they report the performed activities.
From the types of data collection methods, 375 (89.1%) respondents know interview as data collection method followed by record review 161(38.2%) and observation 158 (37.5%).

Data management practice
Almost three fourth (74.3%) of the respondents have good data management practice. Significant number of respondents, 394 (93.6%) keep document of the last survey and 334 (79.3%) agree that redundant data affect the quality of data. Fifty three (12.6%) respondents make appropriate decision without timely information and 411 (97.6%) respondents can make report for the collected data by themselves.
Almost all 409 (97.1%) of the respondents use information at their hand for day-to-day activities, out of which 403 (74.3%) HEWs have good data management practice. Incompleteness of the information (6.7%), irrelevancy of the information (5.2%) and inability to use it (4.8%) were the main reasons reported for not using their own information for their daily activities.

Organizational factors of health extension workers
According to their response 134 (31.8%) respondents did not have reference materials in their office. 404 (96.0%) were supplied with registration books. More than half (56.5%) of the respondents obtain pen and pencil from respective health office. Out of the total respondents, 394 (93.6%) have been supervised by their supervisors. More than three-fourth (81.7%) of the participants were supervised at least once in every month and the remaining 46 (11.7%) and 26 (6.6%) respondents were supervised once in every three and; six and above months respectively. Feedback from concerned body was given for 353 (83.8%) respondents (Table 2).

Technical factors of health extension workers
There were no reporting formats for different activities in the 50 (11.9%) respondents' office. From 371 (88.1%) respondent who have the reporting formats in their office, 202 (48.0%) respondents had face shortage reporting formats frequently. The existing reporting formats were too complex or difficult to understand for 105 (24.9%) respondents. The main reasons for its complexity were use of uncommon words/ terms (60.0%), abbreviations (25.7%), and inconsistency of the formats (5.7%), (Table 3).

Factors associated with data management knowledge and practice of HEWs
As shown in the bivariate models, data management knowledge was significantly associated with availability of registration book, availability of reporting formats, supervision, and frequency of supervision. However, in the multivariate logistic regression analysis, data management knowledge was associated significantly only with availability of reporting formats and frequency of supervision.   In the same way bivariate logistic regression analysis for data management practice was fitted and availability of registration book, reporting formats, feedback and availability of reference material were significantly associated. However, in the multivariate logistic regression analysis, data management practice was significantly associated with availability of data registration book and reference materials with [AOR=5.66, 95% CI: (2.019-15.874)] and [AOR=1.897, 95% CI: (1.178-2.968)] respectively (Table 5).

Discussion
This study attempted to assess data management knowledge and practice of both urban and rural HEWs and its associated factors in Gamo Gofa Zone, southern Ethiopia. The overall data management knowledge of 58.2% HEWs was good according to operational definition set for measurement of data management knowledge. In this study data management knowledge was assessed by measuring the competence of HEWs for collecting, processing and reporting data.
This research showed that significant number (95.7%) of the respondents used data at their hand for their daily activities, and 368 (87.4%) of the respondents made decisions with having timely information. 92.2% respondents keep the documents of the last survey in their office to utilize the information in the future. In general, 74.3% of HEWs were categorized as having good data management practice based on the operational definition set to measure data management practice. 394 (93.6%) respondents were supervised, of which 322 (81.7%) were supervised at least once in every month which is better compared with the previously reported findings from north Gondar Zone (34.7%) [9]. Another survey indicated that two-third of the health posts were supervised during the three months preceding the survey [10]. Feedback was given on the data reported to higher levels for 353 (83.8%) respondents which is higher than the findings of study in north Gondar Zone, Ethiopia (12.2%) [9]. The possible reasons for this difference might be the provision of motor cycles for supervisors in the current study.
The existing reporting formats were complex or difficult to understand for 105 (24.9%) respondents. This finding is in line with assessment of health management information system [11] conducted in Addis Ababa health Bureau (25%). In this study, frequency of supervision and availability of reporting formats were found to be the crucial factor in predicting data management knowledge status of HEWs. Thus, HEWs who had reporting formats in their office were 2.6 times more likely to report good data management knowledge than those who had no reporting formats in their office [AOR= 2.631, 95% CI: (1.331-5.203)]. This may be because those who had reporting format in their office are expected to report their findings after performing other data management tasks such as collection, processing and analysis of their data.
The finding revealed that those Health Extension Workers who were supervised every three months were 4 times [(AOR=4.204, 95% CI: (1.372-12.885)] more likely to have good knowledge than compared to those who were not supervised. Other supervision frequencies such as every month, every six months and every one year were not significantly associated. That means those HEWs who were supervised once in every month, 6 months and every year have no significant difference in their data management knowledge compared with no supervision at all. The possible reason might be the supervision type in every three month was more in-depth with knowledge acquiring   This study showed the importance of the availability of data registration books and reference materials for good data management tradition. Health Extension Workers who had registration book were about six times [AOR=5.66, 95% CI: (2.019-15.874)] more likely to perform good data practice. Also, those HEWs who reported they had reference material on how to manage data were almost two times [AOR=1.897, 95% CI: (1.178-2.968)] more likely to have good data management tradition compared with those who had not reference materials. This might be due to continuous supply of data management recourses to the health extension program which may result in increased utilization of the resources.
The challenges for this study were limited previous publications or validated data collection tools, and lack of defined competencies in data management knowledge and practice for HEWs. To overcome this extensive discussions have been made with HEWs, their supervisors and program managers. In line with this data management practice questions were assessed only by self-administered questionnaire with its associated limitations.
In conclusion, data management knowledge was found to be low whereas, data management practice was found to be high in the study area. The main determinants of data management knowledge and practice were mostly organizational and technical factors. The major factors identified include availability of reporting format, reference books and registration book and also frequency of supervision. Therefore, giving training for HEWs on data management to improve knowledge based work practices, modifying or briefing formats, improving frequency of supervision and provision of adequate and sustainable data management resource is mandatory. Further researches are needed using tools having greater observational components to assess the practice of data management.