Study design and area
Community based cross-sectional study was conducted in rural kebeles of Jimma zone, from March 10 to 30, 2020. Jima Zone is located 350 km away from Addis Ababa, capital city of Ethiopia. There are 21 Woredas in Jimma zone. According to 2007 Ethiopian census report, the total population of Jimma zone was 2,486,155 of which 2,348,487 (94.46%) were rural residents. Among the rural residents 129,939 (5.53%) were elderly population. In terms of household the zone has 521,506 households and 485,214 of them were found in rural kebeles.
Sample size determination and sampling procedures
Sample size was calculated using single population proportion formula with the following parameters; proportion of health care seeking behavior by 57.9 % taken from study conducted in Wolaita zone(12), 95% confidence interval, 5% margin of error, design effect of 1.5, and a nonresponse rate of 5%. Finally, using these parameters the sample size found to be 590.
Stratified sampling technique was employed. Firstly, Woredas in Jimma zone were stratified based on their access to basic infrastructure. The stratification was done by Oromia regional government based on their access to basic infrastructure. Accordingly there were three categories of Woredas namely; woreda-A, woreda-B, and woreda-C. Woreda-A has three 3 woredas(Kersa, Mana, and Agaro), woreda-B ha 6 woredas(Seka chokorsa, Sokoru, Shebe sombo, Dedo, Mencho, Gomma), and Woreda-C has 12 woredas (Limmu kossa, Gumay, Omo nada, Omo beyam, Gera, Tiro afeta, Botor tollay, Limu seka, Sigmo, Sentema, Chora botor, Nono beja). Agaro town was purposively excluded from the sampling because since it was town administration.
Totally 6 woredas were selected from the three categories. From woreda-A, Kearsa, from woreda-B, Dedo and Mancho and from woreda-C, Cora botor, Limmu kossa, and Tiro afeta woredas were selected from the respective strata of woredas through simple random sampling technique. Finally, 18 kebeles were randomly selected from these woredas.
Census was conducted for three days to identify the number of elderly people in selected rural kebeles. Consequently, 5345 rural elderly were identified. Among 5345 elderlies identified 3545 had history of sickness within last 6 months of the census time. The total sample size (590) was proportionally allocated to rural kebeles and participants were selected randomly.
The interval value (k) to employ the systematic sampling was obtained by dividing the total number of elderly people in the selected woredas to the total sample size (3545/590 = 6). The first household was selected through lottery method, then subsequent households were selected with interval of every 6 household. Elderly people of both sex were interviewed.
If there were more than one individual who met the inclusion criteria in a given household lottery method was used to select study participants for interview(12). Elderly (age 60 and above) of both sexes who have experienced any health problem within the past 6 month before the survey were included in the study. However, elderly with the following conditions were excluded; severe hearing impairment, a previous diagnosis of dementia and those who did not encounter any health problem within the past 6 month prior to survey. If selected household did not meet inclusion criteria the next nearby household was considered.
Variables and measurement
The dependent variable was health care seeking behavior of elderly people. The independent variables were socio-demographic factors (age, sex, marital status, occupational status, educational status, income, family size, religion, and ethnicity), individual related factors (living arrangements, dependency, housing condition, nutritional status, family support during illness, health insurance coverage and distance from health facilities), perception related factors (perception toward health service, perceived therapeutic options, perceived price of health service, perceived approach of health professional, perceived severity of disease), awareness related factors (awareness about disease, awareness when to visit health center, awareness on chronic illness) and self-care related factors (feeding style, physical activity, self-medication, traditional medication, and spiritual treatment). Semi structured interviewer administered questionnaire adapted from related studies was used to collect the data. Health care seeking behavior was measured with 5 items in 4-point likert scale. The overall mean value ranges from 5 to 20. Health seeking behavior was said to be poor if the score was less than mean score (< 12.80) and it was said to be good if the score was above the mean score (≥ 12.80)(17).
Data processing and analysis
Twelve individuals who had diploma in nursing were employed as data collectors and six individuals who had Bachelor of Science in health science were employed as supervisor for data collection process. Training was given for all data collectors and supervisors for two day on how to interview and record the data prior to actual data collection period. In addition, to assure quality and reliability of the data pretest was done 5% of the sample size among elderly people who were not actual study participants living outside the selected woredas and kebeles of jimma zone. The data were checked manually for completeness and consistencies on daily basis till the end of data collection. The data were entered into Epi data version 3.1 and exported to SPSS version 23 for analysis. Descriptive statistics was used to summarize the data. Bivariate and multivariate logistic regression analysis were done to identify factors associated with health care seeking behavior. Explanatory variables with p value ≤ 0.25 during Binary logistic regression were candidate for final multiple logistic regression model. Odds ratio (OR) with a 95% confidence interval (CI) was used to measure the strength of statistical association between the dependent and independent variable. Statistically significance was declared at P value ≤ 0.05 in final model.
Ethical considerations
This study was conducted in accordance with declaration of Helsinki. Ethical clearance letter was secured from the institutional review board of Jimma University, Institute of health. A formal letter of permission and support were obtained from each concerned zonal, woreda and kebele administrative bodies. Written informed consent was taken from each study participants. Participant’s right to participate and withdraw from the study was kept. Moreover, to assure the confidentiality of participant’s information, individual identifiers like personal name was not obtained and data were not disclosed to any third party.