Study design and setting
An institutional based case control study was conducted in three public health facilities of Arba Minch town, Gamo zone, Southern Ethiopia from February 20 to March 20, 2019. Arba Minch town is the capital city of Gamo zone with 3 public health care facilities (one general hospital and two health centers) and 33 private clinics (18 level one clinics and 15midium clinics). All 3 public health care facilities are giving ART clinic services for seropositive peoples in the town and surrounding areas.
All HIV positive reproductive age group women (15-49 years) who had ART follow up in all public health facilities of Arba Minch town were source population and those women who fulfilled the inclusion criteria during the data collection period were study population.
All HIV positive reproductive age women who were using long acting contraceptive methods were considered as cases while all HIV positive reproductive age women who were not using contraceptive methods were considered as controls.
All HIV positive reproductive age women who use short acting contraceptive methods and women who were severely ill during data collection period were excluded from the study.
Sample size determination and Sampling technique
The sample size was determined by using the double population proportion approach using Epi Info version 7.02 statistical software package with the assumption of 95 % confidence level (Zα/2= 1.96), 80 % power. Case to control ratio of 1:2 and the sample size was calculated by taking different factors from two different studies done in Bahir Dar North West Ethiopia (11,13).
The larger sample size which was 344 taken from previous study done in Bahir Dar City, Ethiopia was taken for this study(11). After adding 5% for non-response rate, the total sample size was 362(121 cases and 241controls).
Cases and controls were selected from each health facilities that are giving ART clinic services for Arba Minch town and surrounding areas. The sample was proportionally allocated depending on their 3 month performance before data collection. Using systematic random sampling techniques the required sample size were drawn as follows. From Arba Minch General hospital 272 (91 cases and 181 Controls), from Sikela health center 86 samples (29cases and 57 controls) and from Secha health center 3 samples (1case and 2 controls) were included in the study.
Operational Definitions and Definition of Terms
- Long acting contraceptive methods: Modern contraceptive methods that prevent unintended pregnancy for more than 1 year which include Long Acting Reversible contraceptive Methods (LARMs) such as Intra Uterine Devices (IUDs) and sub dermal Implants and permanent contraceptive methods (Tubal ligation) (11).
- Myths heard: When women had ever heard any rumor or misconceptions about LAPMs (13).
- Past experience for LAPMs: It is when a women had ever used LAPMs before current used method (13).
- Current use of contraceptive method: Referred to respondents who responded positively for use of at least one type of contraceptive methods at time of the survey to delay or avoid pregnancy(14).
Data collection tool, method and procedure
Data was collected using structured questionnaire to interview all eligible women and data extraction. It was adapted from previous study and some adjustment were made for this study (15). The questionnaire mainly addressed socio demographic variables of mothers (age, women’s educational status, husband educational status, marital status, wealth index and place of residence), Reproductive health variables (parity, gravidity, number of live children, discussion with husband, discussion with health care providers, family planning counseling, pregnancy intention, experience of contraceptive), Medical history variables (clinical WHO staging, disclosure of HIV status, CD4 count, time of HIV diagnosed and HAART user) and Risky sexual behaviour variables (Sexual intercourse in the past 6 months, using condom, having multiple sexual partner). Five nurses and one public health officer participated as data collectors and supervisor, after training and orientation had given for two days. The principal investigator checked the completeness of the data each day.
Data quality control
Data collectors and the supervisor were trained for two days on data collection tool and the procedure by the principal investigator. Pre-test was done in Chencha hospital one week before the actual data collection, and few amendment were made after the pre-test. The required data were collected after obtaining ethical clearance from Arba Minch University Institutional Review Board (AMU/IRB). In addition permission from the three health institutions and consent from the patient was taken. The entire questionnaire was checked and reviewed for completeness and consistency every day by principal investigator before data entry.
After the data collection, data was coded and then entered, cleaned and edited by Epi Info version 7 and exported to Statistical package for Social Sciences (SPSS) software version 23. Wealth index of the participants was analyzed by principal component analysis method. On binary logistic regression analysis, a variable with P-value of ≤ 0.25 was used as a candidate for multiple logistic regression analysis. Multicollinearity was checked by using standard error greater than two. Multi variable logistic regression with backward method was done to find determinant factors. Odds ratio, 95% confidence interval and P-value <0.05 were used to determine the significance and strength of association with dependant variable.
Ethical approval and clearance were obtained from an ethical review committee of Arba Minch University, College of Medicine and Health sciences. Further permission was obtained from all public health facilities. Confidentiality was maintained by making the data collectors aware not to record any identification information found.