Background
Menarche stands out amongst others as a crucial developmental stage in the lifespan of every normal young girl. It marks the onset of menstrual flow in the life of a developing young girl and comes with many fertility and reproductive implications. Though a commonly overlooked indicator in public health, menarche remains the key developmental marker of a girl’s transition from childhood into adulthood [
1,
2]. It is described as an essential indicator of the physical, nutritional, and reproductive health status of every developing young girl [
1,
3]. Puberty in young girls begins with thelarche (development of breast tissue), followed by pubarche (development of pubic hair), then menarche [
4].
The mean age at menarche in Cameroon varies from urban to rural setting as was determined by Pasquet et al. using the “status quo” method in 1999 [
3]. The study reported a mean age at menarche of 13.18 ± 1.08 years in the urban setting, 13.98 ± 1.55 years in the suburban setting, and 14.27 ± 1.65 years in the rural setting [
3].
Over the last three decades, the age at menarche has experienced a consistent drop [
2,
5]. Epidemiological evidence reveals multiple psychosocial and public health challenges that are due to the declining age of menarche. Early onset of menarche has been associated with early marriages [
6,
7], premature parenthood, breast cancer [
8,
9], development of myoma [
10], psychosocial disorders [
11,
12], metabolic syndrome (diabetes, hypertension, and obesity) [
13‐
16], short stature [
17], preeclampsia [
18], poor academic performance and substance abuse [
2,
19]. A substantial body of evidence from developed and developing nations agrees that early menarche (generally defined as menarche before the age of 12 years) increases the likelihood of adverse sexual and reproductive health outcomes including early pregnancy and childbearing, sexually transmitted diseases, early sexual initiation, and sexual violence among adolescent girls [
1,
2,
20].
The age at menarche according to multiple epidemiological surveys has shown significant variations with economic status, educational attainment, race and ethnicity [
21], place of residence (urban/rural) [
3], nutritional status, and family sizes [
2]. It is therefore a sensitive marker of many population indices including diet patterns, socio-economic status, geographical location, and environmental conditions.
Apart from the two studies by Pasquet et al. [
3] and Monebenimp et al. [
22] in 1999 on age at menarche, Studies on menarche and factors associated with menarche are still very few in Sub-Saharan Africa and particularly in Cameroon. The socio-economic and demographic changes over the years are likely to have been accompanied by changes in the age at onset of menarche. We therefore designed this study to update data on the current age at onset of menarche and to evaluate knowledge of young school girls on puberty, menarche and the menstrual cycle in a rural and urban setting in Cameroon, while doing a comparative analysis of the mean age at menarche across different groups.
Methodology
Study design
A school-based cross-sectional study was conducted in two regions of Cameroon from February to March 2017 targeting young adolescent female students aged 9–16 years. Data collected from these young girls (in Yaoundé and Bamougoum) were going to help us update information on menarche in urban and rural Cameroon. This was done using a pretested auto-administered questionnaire and the collected data analysed using Epi-Info version 7.2.2.16.
Setting
The study was conducted in Yaoundé (urban) and Bamougoum (rural). Yaoundé is the administrative capital of Cameroon and is principally made up of the Mfoundi division (In the Centre Region of Cameroon). A cosmopolitan population characterises Yoaunde with about all the ethnic groups in Cameroon represented. Bamougoum on the other hand, is a rural area in Rural Bafoussam, Mifi division, West of Cameroon. The population is dominated by the Bamilekés (ethnic group) who are generally farmers and small traders with their closest urban centre being Bafoussam central town.
Sampling
A multistep cluster sampling method was used in this study. Among the seven sub-divisions that make up the Mfoundi division, two were randomly selected for the survey (the names of the subdivisions were written on separate pieces of papers and folded, mixed and twoselected without replacement). All secondary schools in each of the two selected divisions were censored, their names written on separate pieces of papers, folded and mixed. One school was then selected randomly from each sub-division. The schools selected were Government Bilingual Practicing High School (GBPHS) Yaoundé and Government Bilingual High School (GBHS) Etoug-Ebe; Yaoundé. GBPHS and GBHS Yaoundé enrolled about 4800 and 6000 students respectively in the 2016/2017 academic year. As concerns the rural setting, due to the small number of schools, from the list of all secondary schools in Bamougoum, two were randomly selected. This was done by writing the names of all the schools on separate sheets of papers which were folded and mixed.. By this process, GBHS Bafoussam Rural and GBHS Nguache were selected. GBHS Bafoussam Rural and GBHS Nguache had an enrollment of 2500 and 3000 students respectively in the 2016/2017 academic year.
At the level of the selected schools, given that our study targeted female students aged 9–16 years (adapted age range for the status quo method of evaluating the mean menarcheal age) who had experienced menarche [
2], based on information from the school registers, Form three to lower sixth classes were included in the study in both urban and rural settings. All female students within this class range who were below 9 years or above 16 years of age or who had not yet experience menarche were not included. In each form or level, if multiple classes existed, one class was randomly selected and included in the study. In a selected classroom, all eligible participants who accepted to participate were enrolled in the study.
Procedure of implementation and data collection
When the protocol and data collection tools were ready, administrative authorisations were obtained from the selected schools and ethical clearance obtained from the institutional ethical review board of the University of Yaoundé I. The data collection tools were pretested in a sample of 15 students in a secondary school in Yaoundé (Faith Comprehensive Secondary School). A total of thirty (30) surveyors were recruited and trained on the consenting and data collection procedure in two training sessions of four hours each and then divided into six (6) teams each made up of five (5) members and headed by a team supervisor.
In each school, every selected class was assigned a team headed by a supervisor. Introduction and briefing on the topic of research was done by the supervisor of each team after which all eligible participants were given the information notice of the research to allow the students obtain written and signed consent of their parents/guardians back at home. All willing participants that had obtained the consent of their parents/guardians to participate were included in the survey after obtaining their verbal assent. The questionnaires were then submitted to each participant to fill with a clear notice of liberty to address questions to the research team members only. The participants were given as much time as they needed to complete the questionnaire, and all questions for clarification on the questionnaire answered. Data collected included; their Socio-demographic data, knowledge on puberty and menstruation, the onset of menstrual flow, and information on the age at menarche. During the submission of the filled questionnaire, the body weight (in kg) of each participant was taken using a digital weighing scale and registered in the questionnaire. During weight measurement, each participant removed her shoes, and any other dress she wore on her uniform (only the school uniform, which is very light in our setting, was used during weight measurement). The height (in meters) of each respondent was then taken in an erect position using a graduated height measuring device (see the questionnaire: Supporting information S
1).
Data management and analysis
All filled and submitted questionnaires were cross-checked for validity. Questionnaires that lacked vital information like the age of the participant and the age at menarche were excluded. Data were entered into a predesigned data entry sheet developed with the statistical software; Epi Info version 7.2.2.16. Proportions and their 95% confidence intervals were presented for categorical variables (occupation, level of education, religion … etc) while the mean or median was presented for continuous variables where applicable (age of the participant, age of menarche, and the number of household occupants … .etc). A participant was classified to have good knowledge if she had a 50% correct score on the questions evaluating the level of knowledge on puberty, menarche, and menstruation. The age at menarche was scategorised into two groups based on the median age at menarche registered in the study.
With a statistically significant threshold set at p-value = 0.05, and dependent on the result of the Bartlett’s test for inequality of population variances, different means were compared using either the T-test or the Mann-Whitney/Wilcoxon Two-Sample Test (Kruskal-Wallis test for two groups) where applicable. As concerns factors associated with menarche, the odds ratio and its 95% confidence interval were used as a measure of the strength of association between age at menarche (dichotomously scategorised based on the median age at menarche) and selected covariates. All variables with a p-value less than 0.5 from the bivariate analysis were all controlled one for the others in the multiple logistic regression model with a statistically significant threshold set at p-value = 0.05.
Ethical considerations
Ethical approval for this study was obtained from the institutional ethical review board of the faculty of Medicine and Biomedical Sciences of the University of Yaoundé I. An information notice was given to each participant or their legal representatives. The written consent and verbal assent of the parent/guardian and participant respectively were obtained before data collection. Only willing and consenting participants were included in the survey. Confidentiality and the participant’s autonomy were respected.
Discussion
Menarche is a physiological event occurring during the development of every normal young girl. This study updates data on menarche in a rural and urban Cameroonian setting, goes further to evaluate the knowledge of young adolescent girls on puberty, menarche, and the menstrual cycle while identifying possible factors influencing the age of onset of menstrual flow in these girls.
Regarding the knowledge of our participants, 56.6% of participants had above 50% of the score (good knowledge) concerning puberty, age at menarche, and menstrual cycle. This finding is similar to that obtained in a Ghanaian study in 2016, which reported an average knowledge of 57% on menarche and the menstrual cycle [
23]. The proportion of participants with good knowledge is unsurprisingly higher in this study, given that the target population was university students. Up to 67.20% of participants from rural settings had good knowledge, whereas only 46.00% had good knowledge in the urban setting. This difference may be because the primary source of information in the rural area was the school (51.10%) whereas in the urban participants it was the radio/TV and friends. The higher proportion of rural participants with good knowledge in our study questions the influence of the information source types on the knowledge of the participants. Probably, participants in the rural zone who have the school as their primary source of information were not influenced by diverse ideas in this field. Also, the quality of teaching care administered in the subject matter in both settings might be different. Our study however did not evaluate the quality of education administered to students in this domain, and therefore we cannot be conclusive in our reasoning.
Moreover, the mean age of participants was 15.1 years for rural dwellers and 14.8 years in urban dwellers. This age difference was statistically significant, with a p-value of 0.00 from unpaired t-test. Older girls have a higher propensity of being exposed to information on puberty and menstruation through parents’ education, studies in school, peer discussion, and other media. This might be a possible explanation of why rural-dwelling girls had better knowledge of puberty and menstruation than the urban-dwelling girls. Also, it is possible that our results might report a relatively lower level of knowledge given the nature of the designed questionnaire, the conditions under which the questionnaire was administered, and the age group considered.
The primary source of information on puberty, menarche, and menstrual cycle in the urban participants was the radio/TV (35.26%) whereas, in the rural area, the primary source of information was the school (51.10%). This is expected, given that the urban areas are generally associated with easy access to audiovisual signals. When we look at the proportion of students with good knowledge in rural and urban settings, we can state that the level of knowledge is still relatively low and even lower in the urban setting. Good knowledge has been associated with better attitudes and good attitudes with practices. This level of knowledge among already menstruating girls could be associated with poor practices as far as sexual behaviour and menstrual hygiene are concerned. Readapting or strengthening the course content of these young adolescent girls in their respective schools is indispensable in improving the quality of knowledge of these young girls. Parental home counselling and sensitisation of the young girl is also of utmost importance.
The mean age at menarche in our study was 12.76 ± 1.33 years, with a mean age at menarche in the urban setting (Yaoundé) of 12.48 ± 1.12 years and the rural setting (Bamougoum) of 13.03 ± 1.46 years. When compared, this difference was statistically significant, with a
p-value of less than 0.001. A study carried out in Cameroon in 1999 by Pasquet et al. reported an average age at menarche of 13.18 ± 1.08 years in urban, 13.98 ± 1.55 years in the suburban area, and 14.27 ± 1.65 years in the rural area [
3]. The trend is similar to the trend recorded in this study, given that the age at menarche decreased from rural, through suburban to the urban setting. Compared to the results of Pasquet et al. the mean age at menarche has dropped by 1.24 years and 0.7 years in the urban and rural settings respectively over the last two decades in Cameroon. This drop of the mean age at menarche corresponds to a decrease of 7.4 and 4.2 months every decade in the urban and rural setting respectively. We are however not very conclusive about this decrease as these surveys were carried out among participants with different age limits. This declining trend of the age at menarche has also been reported in population studies carried out in Gambia [
24], China [
25], and Vietnam [
26]. All the above-cited literature presented significant differences between the ages of menarche in urban and rural settings.
This declining trend of the age at menarche is in line with the high prevalence of early menarche recorded in this survey. According to our findings, 14.87% of these young girls attained menarche before the age of 12 years. This progressively early onset of menarche has been described to be associated with the development of multiple gynaecological morbidities like the development of myoma [
10], cardiovascular disease [
15,
27], metabolic syndrome [
14], type 2 diabetes [
16], preeclampsia [
18], and various forms of cancer [
8,
9].
Comparative analysis of the mean age at menarche across different groups showed that the mean age at menarche was significantly influenced by the state of both parents (both alive/ at least one dead), the current age of the participants, the occupation of the parents, frequency of watching television, and body mass index of the participants. With both parents alive, the standard of living of the young girl is likely to be better given the increased likelihood of a better socio-economic status [
2]. According to a study on age at menarche and socio-economic status in Poland, girls from families with high socio-economic status experienced menarche at an earlier age than girls from families with lower socio-economic status [
28]. Also, multiple studies have agreed that higher BMI is associated with an increased likelihood of early menarche [
29‐
31]. Watching television has been described to have a significant influence on the sexual development and behaviour of adolescent girls. It has also been associated with an increased likelihood of early menarche [
32,
33]. Participants with parents having skilled occupations are generally more likely to experience an earlier menarche compared to those whose parents have unskilled occupations. This left us hypothesising that participants with parents having skilled occupations were more likely to have a better socio-economic state compared to others. This could be a plausible explanation for the difference. Studies have however presented contrasting findings; stating that the age at menarche did not depend on the parent’s occupation [
2].
When we evaluated by logistic regression with age at menarche categorised (Menarche before 13 years), we found that only age and urban/rural residence were significantly associated with age at menarche. As already discussed above, urbanisation is by far the strongest factor associated with menarche among young girls. Girls in the urban setting were more than four times more likely to experience menarche earlier compared to their counterparts in rural settings even upon control for multiple variables. The growing urbanisation of the different rural zones in Cameroon is likely to be associated with a very significant drop in the age of first menstrual flow in Cameroon. This will mean higher exposure to early menarche-linked morbidities in the future. Older girls reported a significantly higher mean age at menarche compared to their younger counterparts. Logistic regression results confirmed age as a statistically significant factor associated with age at menarche. The impact of rapid urbanisation over the years and the rapid changes in the methods of child upbringing associated with an increasing prevalence of childhood obesity could explain this [
34].
The level of education of the male parent and BMI were significant factors only on simple logistic regression. These factors have already been reported to be associated with menarche in multiple studies [
2,
30,
31]. These two factors are possible confounders of the urban/rural residence, given that parents in the urban setting are likely to be more educated than their counterparts in the rural setting. Also, city dwellers end up with a significantly larger BMI compared to those in rural settings [
35].
The data herein should be interpreted with care. Information bias was likely to affect our results given that we conducted the study on already menstruating females. The data given to us was therefore retrospective and could be subjected to errors (failure of memory). Moreover, the participants could lie given that some of the questions were personal and did explore their reproductive development. The data collected in this survey does not permit us to evaluate all possible factors associated with earlier onset of menarche. Other possible factors like level of education and occupation of the mother, the wealth index of the family, and nutritional status were not evaluated in this study. Also, the cross-sectional design of the study does not permit us to establish cause-effect relations between variables. The findings in this research are not representative of all urban and rural zones in Cameroon; more extensive studies representing all the rural and urban zones in Cameroon are necessary. In addition, the difference observed between the urban and rural population in this study could be bigger given that the population of city dwellers is generally cosmopolitan (case of Yaoundé) compared to the case of Bamougoum (dominated principally by the Bamilekés; the major ethnic group of the West Region of Cameroon).
Conclusion
Even though young girls in Bamoungoum (rural setting) have better knowledge of puberty, menarche, and the menstrual cycle, the overall level of knowledge in both settings is relatively good. The mean age at menarche of menstruating girls aged 9–16 years is 12.76 ± 1.33 years, with a mean age at menarche of 12.48 ± 1.12 years and 13.03 ± 1.46 years respectively in the urban and rural settings. The mean age at menarche over the last two decades in Cameroon has dropped by 7.4 and 4.2 months in the urban and rural settings respectively. The mean age at menarche depends on some factors like urban/rural residence, state of the parents (both alive/ at least one dead), the occupation of the male parent, age group, frequency of watching TV, and body mass index. Upon control for confounders, urban residence and age below 14 years were significantly associated with earlier onset of menarche.
Parental home counselling and sensitisation, associated with focused reinforcement of their course content on puberty, menarche, and the menstrual cycle is indispensable in stepping up the knowledge of these young girls. Declining age at menarche serves as an essential indicator of the future burden of gynaecological, and cardio-metabolic, morbi-mortality.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.