Study design
An analytic cross-sectional study employing a quantitative method was used for the study. A questionnaire-based survey was conducted from 29th April, 2019 to 10th June, 2019. The survey was carried out on a university campus which was chosen on the basis that university students are a major donor pool for fertility centers and oocyte donor recruitment agencies. Additionally, the university offers the advantage of accessibility to the study population as it had large numbers of undergraduate female students.
Study setting and population
The university campus which is situated in Legon, Accra, has four colleges, nineteen schools, five institutes and eleven centers of learning. It is the largest of the university’s three campuses. Currently, the student population exceeds 40,000; of which 85.4% are undergraduates. The main campus consists of 14 halls of residence and 2 hostels. All the halls of residence house both sexes, excluding two halls that each accommodate only males and only females.
The target population for the study consists of female undergraduate students between the ages of 18-30 years. This group was selected on the basis that most fertility centers prefer this age group for oocyte donation as they are in their peak fertility years [18] and able to provide informed consent. Additionally, high IVF success rates have been associated with the use of oocytes from younger patients. Further, it is believed that the limited financial resources coupled with a greater financial burden associated with university education makes the financial compensation offered to oocyte donors a valuable source of income for the students [19].
Sample size estimation and sampling
The sample size was determined using the formula for estimating a single population proportion. The proportion for substitution in the formula (23%) was based on a previous study on the willingness of female college students to donate an oocyte when given the opportunity [17]. A precision of 0.05, a confidence level of 95% and a non-response rate of 10% were assumed and a final sample size of 300 was computed.
A multi-stage sampling was employed. The first stage involved selection of the residential halls. A list of all halls of residence with female undergraduates was obtained from the university’s computing system. This yielded 14 halls of residence from which 12 were selected to participate in the study. Of the two halls that were excluded, one hall each provided accommodation to only male students and only foreign students respectively. Each hall formed a stratum from which the survey respondents were selected.
The second stage involved the selection of survey respondents. The total population of female undergraduate students in each of the selected halls was obtained with the aim of selecting equal numbers of female students (25) to obtain a total of 300 respondents. In each hall of residence, a list of all the room numbers with female occupants was obtained. The numbers were coded and a computer-generated random sampling of 10 rooms was obtained. Students from the rooms selected were approached with the questionnaires and those who consented to participate were given questionnaires. The selected rooms were visited one after the other until the total number of 25 students were obtained. The process was repeated in all halls to attain the required samples size.
Data Collection
Structured questionnaires were administered to eligible students who consented to participate in the study. The questionnaires consisted of closed ended, open-ended questions and Likert type questions. The questions were divided into four main sections namely: socio-demographic characteristics (7 items), knowledge about egg donation (11 items), attitude towards egg donation (18 items), and personal choices (11 items). Questions in the sections on knowledge about oocyte donation were adapted from [17] and Bernsen [20], whereas items in the sections on attitude towards oocyte donation and personal choices were adapted from [21] and [17].
Twelve research assistants, each representing a hall of residence, were trained in data collection procedures and good ethical conduct. A pretest was conducted in a different tertiary institution to ascertain clarity of the questions and anticipate any practical difficulties. Data collection took place in the halls of residence, mostly in the late afternoon, evenings and weekends as most students were available at these times. Returned questionnaires were scrutinized to eliminate errors and ensure completeness prior to data entry and analysis. Questionnaires with several missing entries were eliminated.
Study variables and data processing
The responses to the question “Will you be willing to donate oocytes (eggs) for in-vitro fertilization (IVF) if you were approached?” were ‘Yes/No’ indicating willingness and unwillingness to be an oocyte donor respectively. This was the outcome variable and it was dichotomous. Demographic characteristics of respondents were age, sex, college of study as a proxy of the program of study, marital status, monthly expenditure in Ghana cedis as a proxy of income levels, occupation and place of residence when outside the campus. Other variables included knowledge of oocyte donation within the context of assisted reproductive technology (ART) and the attitude towards oocyte donation.
Knowledge was assessed as a composite variable made up of seven items. Each question was scored ‘0’ for an incorrect answer and ‘1’ for a correct answer, resulting in a minimum score of zero and a maximum score of seven. To assess the overall knowledge score, a score above three was interpreted as a high level of knowledge and a score of three or less, denoted a low level of knowledge.
Respondents’ attitude, ethical stance and concerns regarding oocyte donation were assessed using 18 questions measured on a 5-point Likert scale, divided into three subscales namely: (i) General attitude towards egg donation (9 items), (ii) Ethical aspects of egg donation such as cash rewards for donors (2 items) and (iii) Issues of anonymity (7 items). A Cronbach’s alpha reliability co-efficient was used to assess the level of internal consistency between the questions within each subscale. Satisfactory values of 0.67, 0.80 and 0.78 were obtained respectively. Among the subscales, ‘General attitude’ was used as an indication of overall attitude towards oocyte donation, with scores ranging from 9 to 45. A respondent’s attitude was categorized as positive if the sum of scores was greater than 24 and deemed negative if a score of less than or equal to 24 was obtained.
Statistical Analysis
Descriptive statistics generated for socio-demographic characteristics of respondents included frequencies, percentages, mean and standard deviation. The Pearson’s chi-square test, Fischer’s exact test and logistic regression analysis were used to determine association between willingness to donate an oocyte or not (outcome variable), and socio-demographic characteristics, level of knowledge about oocyte donation and the attitude towards oocyte donation (explanatory variables). Crude and adjusted odds ratios were presented with the 95% confidence limits constructed around the estimates.