4.1 Prevalence and socio-demographic characteristics

The average age of the participants was 34 years (standard deviation ± 4.95) with a minimum of 21 years. Adolescent girls and young adults accounted for 17.62% of the ICC diagnosed and treated during our study period. 74.35% had their first sexual intercourse before the age of 18 and 34.90% were HIV-positive. A study carried out in California,¹⁵ which included a highly-varied and mixed population of 13,624 cervical cancers with complete data on age at diagnosis, race, ethnicity, socioeconomic level, and marital status from 1996 to 2005, used logistic regression to conclude that more than 58% of ICCs involved women under 50 years of age, with 46% of cases in the delayed phase. Older age, low socioeconomic status, and single woman status predicted the diagnosis of late-stage ICC regardless of race or ethnicity. Some prominent contributing factors outside of race were found in our population, including 40% not attending school, 60% without a fixed income, and 36% living alone which indicated low socio-economic status.

4.2 Mode of admission, symptoms

Patients had been referred without any delay from the provider or referral (38.46%) because they had been referred to the emergency department in a state of hemorrhagic shock (15.38%) or for severe anemia (23%). The common initial symptoms were abnormal vaginal discharge, abdominal pain, abnormal vaginal bleeding including post-coital bleeding (PCB), and inter-menstrual bleeding (IMB). Although more than half (52.38%) of patients had abnormal vaginal discharge as their earliest symptom, abnormal vaginal bleeding was the main complaint (79.50%) of women who saw an HCP for the first time compared to 62.70% in Nepal.¹³ At least 20.5% (8/39) of the women had two consultations with the HCP and the rest used traditional treatment before seeking emergency care. A cervical examination was not carried out in almost three quarters (71.80%) of cases, close to the observed value of 78% in Nepal, and more than a quarter of their symptoms (28.20%) were misinterpreted at the initial consultation with an HCP compared to 90% in Nepal.¹³ These rates could be explained by the fact that patients were consulted at life-threatening and late stages in FCHCs. Patients had an ulcerated cervix in 56.41% of subjects with 79.50% of those showing bleeding as the main symptom at consultation compared to 65.80% in Morocco,¹⁶ whereas 81% of patients were unemployed in Côte d'Ivoire compared to with 80.80% in Morocco. Only 48.72% of the patients were diagnosed at an early stage (IIA) and the rest were diagnosed at an advanced stage (≥IIB) in Côte d'Ivoire as in California and Morocco where 60.1% of the patients were diagnosed at stages IIA-IIIB.^16,17

4.3 Diagnostic path

This study identified different latency periods in cervical cancer diagnosis, with total diagnostic delays derived from patient delay, provider delay, and diagnostic waiting time. Although no standard definition was found for the concept of delay, studies on diagnostic delay have several common themes regarding the length of the delay as a function of the dates of important events in a patient's diagnostic pathway.^12,14,15,18 A similar determination of cancer diagnostic methods was been applied to previous studies.^12,13,15,19 The variation can be found in the point of dichotomization of each type of delay into a long and short delay, but it is highly contextual.^12,20–22 Hansen has categorized delays as short or long according to quartiles and described delays with respect to the median and interquartile range.¹² Other studies have considered some of the periods such as days, weeks, or months for this purpose.^{13,14,21–23}

4.4 Estimates of delays in diagnosis

This study revealed that the median patient delay was 122 days with a long delay in 84.16% of patients compared to only 68 days and 57% of patients in Nepal. Similarly, the median provider delay was 23 days with 79.48% of patients with a short delay compared to 40 days with 80.90% of patients with a long delay in Nepal.¹³ Referral time was found to be comparatively low (average 5 days) with more than two-thirds (89.74%) having a short referral time, whereas in Nepal it was 68.20%. From the first visit to the diagnostic center, the majority (90%) of patients had to wait more than two weeks for the diagnosis of cervical cancer (mean diagnosis waiting time of 31 days versus 66.20% of patients with a median of nine days in Nepal). The mean value of the total diagnosis delay was 301 days with a median of 209 days and a longer total diagnostic delay in 87.18% was more than 157 days in Nepal with 77.30% delay. This diagnostic delay was found to be higher in both Côte d'Ivoire and Nepal¹³ when compared to that of developed countries.^23,24 A high prevalence of long delays in diagnosis of more than three months is extremely unacceptable if the cancer is to be treated at an early stage. The longer the duration of symptoms until a diagnosis is consistent with the high prevalence of advanced cervical cancer in Côte d'Ivoire and Nepal.^10,13 Patient and HCP delay accounted for most of the delays. Indicating a significant delay, Ivorian women had suffered longer patient delay compared to other delays with a wide interval of 8–2256 days as in Nepal.¹³ The longer patient delay in Côte d'Ivoire may be due to the influence of patient characteristics such as high levels of illiteracy, poor health consciousness, poor economic conditions, and problematic health behavior such as ignoring mild gynecological symptoms as well as dependence on traditional health care practices.^10,20 In previous studies, researchers have also revealed that patient abilities for recognizing the severity of symptoms and subsequently ignoring increased patient delay in cervical cancer diagnosis.^20,21 In some populations, common symptoms such as vaginal discharge are not recognized as warning symptoms and in most cases assistance is not sought until it becomes unbearably obvious, ultimately leading to increased patient delay and total delay in diagnosis,¹³ thereby resulting in 38.46% of an emergency evacuation.

This study establishes that delay caused by HCPs is another major delay in the diagnosis of cervical cancer. Although at a low level, provider delay has also been observed in previous studies, even in developed countries.^12,16 The observation of medical delay in Morocco and Nepal, where 61% and 80.94% of patients suffered a delay of more than 30 days, respectively, was lower in Côte d'Ivoire at only 20.52%.²⁵ The low proportion of delay by the HCP is because 38.5% of the patients were consulted in a state of emergency either associated with hemorrhagic shock (15.38% of cases) or severe anemia (23% of cases). HCP delay in Côte d'Ivoire can be discussed from various perspectives such as access to services, education level of HCPs, the existing healthcare system, and policy. In both Côte d'Ivoire and Nepal, the first point of contact such as doctors' surgeries, FCHCs, and private medical NGOs at the community level are managed by HCPs with basic medical training. These health care workers often lack knowledge and skills in gynecological examination and cervical cancer screening. In the existing health system in Côte d'Ivoire, not all women have access to gynecologists or suitable doctors for their gynecological symptoms.²⁶ This argument is consistent with the findings that women have to visit several health facilities before they are finally referred to an appropriate diagnostic center. The structure of the health care system, the referral mechanism, socio-cultural factors, the level of knowledge of HCPs, and the asymmetrical relationship between HCPs and patients influence patient tracing practices.^27,28 Inadequate knowledge of the etiologies of cervical cancer, alarming symptoms, screening, procedures, and treatments among health practitioners further contribute to delays in diagnosis,²⁹ eventually leading to the misdiagnosis of cancer. Failure to recognize the vicious symptoms of cancer and/or not being able to provide a cervical examination via the HCP at the initial consultation creates unnecessary visits to different health institutions.^14,30 Patients who complain of alarming symptoms such as abnormal vaginal bleeding or severe pelvic pain often consulted a FCHC in a shorter time than patients with other more mild symptoms such as foul-smelling vaginal discharge. It was only observed that speculum placement by the care provider is often of less interest in women without vaginal bleeding. The duration of the delay has been reported to be shorter in patients who have a gynecological examination by the HCP for patients exhibiting alarming symptoms.^13,14

The distribution of the length and frequency of each type of delay in diagnosis varied between groups of participants. Older, illiterate, and low-income women experienced longer patient delays, HCP delays, diagnosis waiting times, and total diagnosis delays. Studies have also found that advanced age is a risk factor for delayed diagnosis of cancer, including cervical cancer.^20,21,31 A higher proportion of long delays in some patient groups depict barriers in access to health care and longer delays in diagnosis among this population.²¹ The longer total delay in diagnosis (>90 days) was observed more among patients aged 30 years or older (85.30%), and patients who were illiterate or had only primary education (82.2%). Care provider delay was observed at 87.5% with an average of 126 days among those aged 30–39 years, and 87.5% with an average of 127 days among those with less than secondary education. PSS delay and total diagnostic delay were found in 100% of patients whose cervix was not examined at the initial consultation with an average of 112 days outside of the emergency referral.

4.5 Prognosis

Cervical cancer caused 35.89% of deaths among patients, 85.71% of which occurred in the first year after treatment and 91.66% in stages III and IV. The majority of patients who died (85.71%) were diagnosed with at least stage 2 B according to FIGO as in California.^17,32 Among the deaths, 35.71% had started treatment but could not finish it for lack of financial means (three patients dropped out of the first treatment and one dropped out of the fourth treatment, and only one patient of them underwent surgery without chemotherapy). The causes of death are multiple and can be summed up as late diagnoses where patients were seen at an advanced stage (stage III/IV, FIGO) and did not have access to curative treatments. Such cases represented 70–80% of cases in Abidjan which is worsened by the high cost of treatment (CIC).² The majority of patients 57.14% who died (8/14) were HIV-positive. A more recent study, again in Abidjan, found that 25% were HIV-positive out of 152 cervical cancer patients, compared to 4.70% HIV-negative patients out of 157 total patients without invasive cancer. This study noted respectively 9% cervical cancer, 52% oncogenic HPV, and 3.90% cervical cancer, 33% HPV in HIV-positive and HIV-negative patients.⁵ Respectively, 47.62% and 38.10% of patients were in their first and second year of survival which can be explained by the fact that 76.20% of them were at most at stage IIA. In California,68.75% of patients had benefited from radical treatment compared to 90% of 5-year survivor were diagnosed at stage I.³² In Nigeria, 65 patients with histologically confirmed ICC were followed up with. The majority (72.30%) were diagnosed with advanced stages of ICC. A simple total abdominal hysterectomy was performed in 38.90% of patients who were diagnosed at an early stage of the disease. After a cumulative follow-up of 526.17 months, 35 deaths (54% of follow-ups) occurred with an overall mortality rate of 79.8 per 100 women-years.³³