Cervical cancer is a common cancer among women worldwide. An estimated 528,000 new cases and 266,000 deaths occurred in 2012. More than 85% of invasive cervical cancer cases occur in low- and middle-income countries. Cervical cancer ranks as the most common cancer among women in Ghana. We conducted a retrospective study to assess the descriptive epidemiology of cervical cancer in Ghana. We describe cervical cancer incidence and mortality rates for the regions served by two large hospitals in Ghana.
Patients and Methods
Information for women diagnosed with invasive cervical cancer between 2010 and 2013 was collected from the Komfo Anokye and Korle Bu Teaching Hospitals through review of medical, computer, and pathology records at the oncology units and the obstetrics and gynecology departments. Telephone interviews were also conducted with patients and relatives. Data were analyzed using summary statistics.
Results
A total of 1,725 women with cervical cancer were included in the study. Their ages ranged from 11 to 100 years (mean, 56.9 years). The histology of the primary tumor was the basis of diagnosis in 77.5% of women and a clinical diagnosis was made in 22.5% of women. For the 1,336 women for whom tumor grade was available, 34.3% were moderately differentiated tumors. Late stage at presentation was common. The incidence and mortality rates of cervical cancer increased with age up until the 75 to 79–year age group and began to decrease at older ages. The Greater Accra region had higher overall incidence and mortality rates than the Ashanti region.
Conclusion
Our study suggests that improvements in the application of preventive strategies could considerably reduce the burden of cervical cancer in Ghana and other low- and middle-income countries. The study provides important information to inform policy on cancer prevention and control in Ghana.
Introduction
Cancer causes more deaths in the world than HIV/AIDS, tuberculosis, or malaria.1 Among female cancers, cervical cancer is the fourth most common in the world, with an estimated 528,000 new cases in 2012.2 Despite the fact that cervical cancer is potentially preventable through vaccination and screening, more than 250,000 women per annum are estimated to die of the disease worldwide.2 Incidence and mortality have declined in most high-income countries, mainly as a result of the introduction of cervical screening. However, this is not the case in most low- and middle-income countries, where approximately 85% of the disease occurs.2 The incidence and mortality rates are high in Africa and some parts of Asia, and are low in Australasia and
West Asia.2
In Ghana, cervical cancer is the most common cancer among women, with an estimated 3,052 new cases and 1,556 deaths in 2012.2 Despite the magnitude of the problem, accurate data on cervical cancer incidence and mortality are not available. Currently, there are two hospital-based cancer registries, located in the two large tertiary referral hospitals: Komfo Anokye Teaching Hospital (KATH), Kumasi, and Korle Bu Teaching Hospital (KBTH), Accra. These hospitals are the main referral centers where patients with malignancy are diagnosed and treated, and they see most cases of cervical cancer in Ghana, receiving referrals from all over the country.
The registry at KATH has recently been converted into a population-based cancer registry to cover the city of Kumasi. Ghana is divided into 10 administrative regions. The KATH total catchment area covers approximately 50% of the population of Ghana.3 The KBTH catchment area covers the entire southern part of Ghana, but the hospital also receives patients from other parts of the country. Hence its catchment area is not as well defined.4
Several studies of cancer in Ghana have focused on all cancers and have mainly been institution-based studies without a reference population.5-7 Additionally, no previous research has tried to assess both the cervical cancer incidence and mortality in Ghana and link the data from the two referral hospitals. Knowing the incidence and mortality rates of cervical cancer is important for the formulation of policy and implementation of control measures. This study examined the characteristics of cases of cervical cancer in Ghana and estimated the incidence and mortality rates in two regions of the country.
Patients and Methods
The study population consisted of all histologically confirmed and suspected cases of invasive cervical cancer in the oncology units and the obstetrics and gynecology departments at the KATH and the KBTH between 2010 and 2013. We reviewed paper-based (medical), electronic, and pathology records at the oncology units and the obstetrics and gynecology departments of the two hospitals. We collected information about women newly diagnosed with invasive cervical cancer during the 2010 to 2013 time period.
One hundred twenty-eight deaths from cervical cancer were recorded in patients’ medical records. Most of the women were lost to follow-up at the two hospitals. Hence information on the outcome of disease (whether women were alive or dead) could not be obtained from review of paper-based and electronic records for the majority of women. However, we attempted to contact all women with a diagnosis of cervical cancer by telephone to obtain basic descriptive and outcome information for this study. For those who could not be reached as a result of death or other reasons, their relatives were contacted. For some women with cervical cancer, the hospital did not have any contact information.
For each department, available information for women with cervical cancer was abstracted from paper-based, electronic, and pathology records onto a standard data collection sheet. Once data from all departments were reviewed for accuracy, the following variables were used to link cases between departments: histology report number, patient’s name, age, and telephone number. After linking the data from the two departments and removing duplicates, data were stripped of all identifiers and the final data set was defined for analysis. Because deaths from cervical cancer were only obtained for cases from 2010, only the mortality rates for the 2012 to 2013 periods are presented because most women in Ghana who die as a result of cervical cancer do so within 2 years.
Approval to conduct the study was sought from the University of Otago Ethics (Health) Committee, Ghana Health Service Ethical Committee and the Committee on Human Research, Publication and Ethics of the Kwame Nkrumah University of Science and Technology and KATH, Ghana. The two hospitals also have internal ethical committees that assess proposed research to take place in the hospitals; the research was also submitted to these committees.
Statistical Analysis
We used STATA version 12.1 (STATA Corporation, College Station, TX) for the data analysis. Frequencies were run for all variables for the purposes of describing proportions or percentages. Statistical tests were performed using the χ2 and Fisher’s exact tests. Statistical significance was determined at a P value < .05.
Population data from the Greater Accra and Ashanti regions were obtained from the 2010 Population and Housing Census in Ghana. The census provided information on the population by region of residence, age, and sex.8 The census data were not extrapolated for 2011 to 2013 because little increase in population was expected for that time period. The population data from the Ghana Statistical Service showed that there was a < 1% increase in the population from 2010 to 2013.
The place of residence was obtained from the medical records and was verified during telephone interview for those who could be contacted. The place of residence had not changed for the majority of women (approximately 95%) and was similar to that recorded in the medical records. To calculate the person-years, the number of females in a region (as estimated by the 2010 census) was multiplied by four (the number of years the cases were accrued) for each age group for incidence and by two for mortality.
Cervical cancer mortality rates were also calculated. Death ascertainment was done by telephone interview of relatives and review of medical records. The deaths recorded in the medical records agreed with those obtained by telephone interview when relatives of women were contacted. To provide regional comparisons adjusted for age, the age-standardized rate (ASR; with 95% CI) was calculated using the direct method and applied to the WHO world standard population for each region.9 For crude and age-specific rates, 95% CIs were calculated using the binomial exact method.
Results
The basic demographic characteristics of women diagnosed with cervical cancer (recorded) are shown in Table 1. A total of 1,725 women with cervical cancer were identified for the 2010 to 2013 time period in the two hospitals. The majority of women with cervical cancer were recruited through the records of the oncology units. The mean age at diagnosis was 56.9 years (range, 11 to 100 years). One subject, whose age was recorded as 11 years, was diagnosed with adenocarcinoma of the cervix. However, the type of adenocarcinoma was not stated. Most of the women (58.7%) with cervical cancer were of a trader or farmer occupation.
Recent Comments