Study area and period

This study was conducted in all Goba Town Public Health Facilities (Madda Walabu University Goba Referral Hospital, Goba Health Center and Harawa Sinja Health Center) from September to November, 2022. The study area-Goba town is located in the Bale Zone, Oromia National Regional State, Southeast Ethiopia. Goba town is located at a distance of 430 km away from the capital of Ethiopia- Addis Ababa and 12 km from Robe town.

Study design and population

Facility based multi-center comparative cross-sectional study design was conducted among women attending family planning unit in Goba Town Public Health Facilities. The study population were comprised of women between 15 and 49 years of age, those users of hormonal contraceptives, attending the family planning unit in Goba Town Public Health Facilities for follow up or discontinuation of service, apparently healthy women, who voluntarily consent to participate in the study, fulfill inclusion criteria and selected as a sample (for users). The other study population were comprised non-users of hormonal contraceptives methods attending the family planning unit in Goba Town Public Health Facilities for follow up or discontinuation of service in those users of IUCD or new comers plan to seek any type of contraception, apparently healthy women, who voluntarily consent to participate in the study, fulfill inclusion criteria and selected as a sample of the same age and BMI with users (for non-users).

For hormonal contraceptives users, women between 15 and 49 years of ages who have been using progestin only pills, DMPA and implanon for at least 3 months; and for non-users, women between 15 and 49 years of ages, women who have been using IUCD and new visitors those plan to use any type of contraception were included in the study.

Women with known chronic disease (diabetes, hypertension, liver dysfunction, cardiovascular disorders); pregnant women; women with chronic alcohol and/or tobacco use; BMI ≥ 30 or obesity; and those currently using medication known to affect lipid profile (including corticosteroids, antipsychotics, diuretics, anticonvulsants, retinoid) were excluded by reviewing their medical records and/or by asking. For non-user, women who have been using any hormonal contraceptives during the last 1 year were also excluded.

Sample size determination and sampling technique

The sample size was calculated by using mean estimating formula. The value of the outcome variable was taken from the previous study done in Ethiopia [20]. After calculating all the values of lipid profile components, the component with the highest value (total cholesterol) was taken. The reported mean ± SD values of total cholesterol for users were (183.12 ± 40.56) whereas, (161.76 ± 29.45) for non-users. By assuming; significance level = 95%, 80% power of the test, type of test = two-sided, Z α\2 = the critical value at 95% confidence level of certainty (1.96).

Therefore, the value of sample size calculated was 56. By considering design effect of 1.5 (56*1.5 = 84) because the samples were taken from different institution, the calculated sample size was 84. Since equal number of cases and controls was used, nˣ2 = 84ˣ2 = 168. Taking non-response rate of 10% = 10/100ˣ168 = 18. Then the total sample size was 186 (168 + 18). So in our study, 93 hormonal contraceptive users who fulfill inclusive criteria of case and 93 non-users that fulfill inclusive criteria of control were included.

First, the three public health facilities (MWU GRH, Goba Health Center and Harawa Sinja Health Center) were included. Sample size was proportionally allocated for each health facilities using proportional allocation formula. Then, a systematic random sampling technique was used to select study participants in each health facilities. i.e. every 2nd values was selected according to their order of visiting by using K = N/n formula.

Operational definitions

Castelli Risk Index: are lipid ratios estimated as TC/HDL-c and LDL-c/HDL-c to predict cardiovascular disease risk.

Castelli Risk Index Classification: If the difference in the ratio of TC/HDL-c (Castelli Risk Index I) or LDL-c/HDL-c (Castelli Risk Index II) between hormonal contraceptive users and non-users are < 0.11 (low risk), 0.11–0.21 (intermediate risk), > 0.21 (high risk) for development of CVD.

Contraceptives: birth control methods that include pills, Depo-Provera, IUCD, implant, natural or surgical procedures to prevent conception.

Coronary Heart Disease: chest pain or heart attack due to atherosclerosis which is secondary to excess level of TC, LDL-c and TG.

Hormonal Contraceptive users: Women who have been using progestin only pills, Depo-Provera and implanon.

Lipid Profile: a panel of blood test of TC, TG, HDL-c and LDL-c.

Non-hormonal Contraceptive users: Women who have been using IUCD, natural methods and surgical method.

Data collection tools and procedures

Socio-demographic characteristics, type and duration of contraceptive use was collected from the participants by using Afan Oromo and Amharic version of the semi-structured questionnaire by trained midwifery professional. Five milliliter of venous blood was collected using sterile syringes and needles from each participant using aseptic procedure by laboratory technologist. Then, the blood samples were left for 30 to 40 min at Room Temperature, centrifuged at 3000 RPM for 5 min and serum of blood samples were separated. Estimation of lipid profiles (TC, TG, LDL-c and HDL-c), were conducted using Bio-system mindray BS-200 Chemistry Analyzer. TC/HDL-c ratio (Castelli index I) and LDL-c/HDL-c (Castelli index II) were calculated to determine the CVD risk. Weight and height was measured by using weight scale and stadiometer respectively. BMI was calculated from weight in kg/ height in m2.

Data entry and statistical analysis

Data entry was done using Epidata software version 4.6.0.2. SPSS software (version 25; for windows) was used for data analysis. After complete entry of all the data, soft copy was checked with its hard copy to see the consistency. Standard statistical methods are used to determine the mean, standard deviation (SD) and the range. Student's independent t-test was used to compare the results of lipid profiles of hormonal contraceptive users and control group. One-way ANOVA (analysis of variance) was used to identify the variation of lipid profile between progestin only pills, DMPA and implant users. Simple linear regression was used to determine the change in lipid profile in relation to the duration of hormonal contraceptive use. The normality and homoscedasticity of the data were checked by using the kolmogrov-smirnov test and Levene’s test respectively. Variables with P-value < 0.05 was considered as a statistically significant. Finally, the result was present using graph, table and text.

Data quality control

Training on the contents of the questionnaire, data collection techniques, and research ethics was given for data collectors. The quality of the data was controlled carefully. Close supervision was done by the principal investigator throughout the data collection. Collected data checked for completeness and consistency daily. The completed questionnaires were rechecked repeatedly by the principal investigator to maintain the quality of data. Pretest of the questionnaire was conducted in 5% study participant at Robe General Hospital for 1 week prior to actual data collection to check tools reliability.

Ethical consideration

Ethical clearance was obtained from Jimma University, Institutional Review Board (Ref. No: 73/22); a support letter was obtained from the Health Research and Post Graduate Director’s Office for each study area and concerned health facilities was communicated with formal letter (Ref. No: H/I/R/P/G/D/63/2022). Written consent was obtained from each study participant. Each study participant was informed about the research, and confidentiality of information was maintained during data collection, analysis, interpretation and publication of results. Findings were communicated in aggregated form and individual information was kept confidentially. The culture, religion and society values were highly respected in every concern during data collection. COVID-19 prevention protocols were taken for both data collectors and respondents during data collection. Clients with elevated lipid profiles were linked for further investigation and treatment.