How accurate are India’s cancer registries? | Explained

Premium

The story so far: More women in the city of Aizawl are diagnosed with lung cancer than in all of Asia. Breast cancer cases are swelling in Indian metropolitan cities, while the numbers of cancers of the mouth and tongue are higher throughout Bhopal and Meghalaya’s East Khasi district.

(For top health news of the day, subscribe to our newsletter Health Matters)

This is data from India’s cancer registries, maintained under the National Cancer Registry Programme (NCRP). These registries function as the data storage backbone for a nationwide cancer surveillance programme, recording information about what types of cancers occur and where, and what about a region’s demography makes some cancers more common than others.

ALSO READ

The anatomy of women’s cancers: how does gender shape diagnosis, treatment?

But they present an incomplete picture, covering only about 16% of India’s 140 crore population — less than the number of people living in Uttar Pradesh. The registries suffer from a lack of human and financial resources, are concentrated in urban regions, and the data collected is often delayed and inaccurate. Low awareness, high cost and pervasive stigma further eclipse the true picture of cancer’s footprints in India.

According to some experts, India is poised to face a ‘tsunami’ of chronic diseases like cancer, with cases expected to cross the 15 lakh mark by 2025. But data from these registries may not suffice to tackle, or even understand, this wave.

How does India collect cancer data?

Since 1982, the National Cancer Registry Programme sources records via two mediums: population-based cancer registries (PBCRs) and hospital-based cancer registries (HBCRs). PBCRs trace incidence and trends for a defined geographical population in a region, while hospital-based registries focus on the medical symptoms, diagnosis and care of cancer cases. NCPR keeps records of 38 PBCRs and 189 HBCRs distributed through India; its latest report covers 28 PBCRs and 58 HBCRs owing to the quality and rigour of data collected. This marks an improvement from the first 20 years, when about six HBCRs and 13 PBCRs maintained records for inclusion.

These registries are regulated under the National Centre for Disease Informatics and Research (NCDIR), of the Indian Council of Medical Research (ICMR).

Source: The National Cancer Registry Programme report 2020.

Since cancer is not a notifiable disease in India, data collection is active in nature: trained staff are required to periodically visit and interview different sources: hospitals, nursing homes, clinics, diagnostic labs, vital statistics departments, hospices and registrars of births and deaths, per the NCRP website. Personnel use a standardised core form to profile the patient and the disease and follow up with death certificate notification. Data then undergoes quality control and duplication checks (only patients who have resided in the area for a minimum period of one year are eligible) and is further segregated into sub-categories like cancer types, age and sex.

Why is cancer registration important?

Both the nature of cancer and the principles of disease management demand strong surveillance systems. Studies show cancers in low- and middle-income countries (LMICs) are diagnosed at a later stage, making treatments complex and expensive. Only 67.4% of cancer deaths are registered in India, a conservative estimate of cancer mortality, experts say. By recording cancer trends over time, functional registries can guide diagnostic efforts, prioritising prevention and screening campaigns, particularly in areas with low access to healthcare facilities. Time is of the essence with cancer management, Mridu Gupta, founder of CAPED India, a cancer advocacy group, told The Hindu earlier. Well-maintained registries minimise the chances of delays, oversights and exclusions.

Operational registries also act as pole stars for evidence-based cancer management. For instance, the NCRP data indicates that a woman from Bengaluru is more likely to be diagnosed with cervical cancer as opposed to someone from Manipur, where lung cancers were more common among women, an insight that can influence health screening and treatment decisions. Registries offer a blueprint on how to build defences around cancer as a disease; data from the Census and the National Family Health Surveys have similarly helped experts understand the prevalence of other health crises, such as anaemia and malnutrition.

A parliamentary committee in August urged the Health Ministry to expand coverage of population-based registries to rural areas to help with “data-driven and evidence-based policy formulation” (currently less than 1% of the rural population is included in PBCRs).

ALSO READ

Data | India’s Cancer Crisis: Rising Prevalence of the Most Expensive Disease to Treat

Moreover, the alarming rise of non-communicable diseases is also laying bare health inequities baked into the system. From diabetes to cardiovascular diseases, diagnosis and treatment remains a challenge for people belonging to disadvantaged genders, castes, classes and ethnicities. A new Lancet Commission report found that more than 67% of cancer deaths among Indian women were preventable; gender inequality and restrictive norms resulted in health systems underestimating the cancer burden and subsequently deprioritising screening, prevention, research and treatment efforts.

Patchy records translate into exclusionary disease planning, and cancer registries are seen as a way to trace inequities before they congeal into stark mortality figures. Cancer registration lags globally, unevenly covering only approximately 21% of the world population; only one in five LMICs have the data to drive cancer policy, says the International Agency for Research on Cancer.

The challenges today

Poor funding ails both infrastructure and personnel needs. A 2020 study by Amiya Bhatia et al, which assessed the functioning of six PBCRs in urban and rural areas, cited lack of funding as the primary challenge, which in turn affected staff salaries, retention and ability to travel long distances to contact patients.

Due to the paucity of trained people, Indian registries often rely on medical records or operation books, which may prove insufficient both in terms of quality and quantity of data. The PBCRs which did not interview patients, as is the case with most urban PBCRs and HBCRs that rely on hospital medical records, were limited to sociodemographic components. While some PBCRs collect information on income, education, occupation, marital status and religion, most do not publish it for further analysis. “No PBCR collected information on ethnicity, scheduled caste, scheduled tribe or caste,” the 2020 study also found.

Moreover, since the patient is required to be a resident for more than a year to be eligible, cancer trends among migrant populations are left out. Follow-up of patients to understand the quality and duration of survival is also not a practice integrated into the PBCR ecosystem. This breeds duplication and blind spots: name changes due to marriage or other reasons, spelling mistakes, and patients visiting multiple facilities for separate opinions may further complicate matters.

Reports over the years have flagged the quality, “accuracy and detail of information on cancer diagnosis, and timeliness in updating the registry databases.” There is a lag in real-time data collection: some registries recorded data several months after diagnosis (about 800 cases from a private hospital in New Delhi could not be matched with Delhi’s registry due to a backlog). A lack of linkages between cancer registries (such as a hospital-based registry and population-based) may also leave people out: data of a cancer patient may not be recorded if they live outside the defined areas of a PBCR, for instance.

The NCRP is piloting a cancer atlas project in Punjab and Haryana that visualises a geographical map of the disease. The project aims to pivot to online data entry, provide training to personnel and strengthen the pathology departments (in medical colleges and hospitals that are points of diagnosis for over 85-90% of cases, according to NCRP. The project aims to build a “network of computers linked to a website so that the information (mainly patient identification details including area of living, and site and morphology of tumour) on all malignant cases would be transmitted as and when they are reported in the department of pathology.”

The Parliamentary panel in August also suggested that PBCRs can be linked to the Ayushman Bharat Digital Mission to record real-time data of cancer patients.

While researchers and practitioners call for efficient ways to interlink data, a 2018 paper simultaneously cautioned about the confidentiality and privacy aspects of cancer registration in India.

India’s methodology also differs vastly in comparison to global standards. A 2018 paper found registries in Delhi, Mumbai and Trivandrum to be “of lower quality” as they did not follow the IARC quality indicator method of Surveillance Epidemiology and End Results (SEER), seen to be a gold standard. The NCRP uses data from the 2011 census to estimate mid-year population estimates; the Global Cancer Observatory in comparison relies on United Nations population estimates.

Lack of awareness about the incentives, benefits and need for cancer registration also plagues the process. An analysis of NFHS data shows women’s deaths are seldom registered, along with other disadvantaged groups such as rural residents, economically poor groups and deprived caste communities, which was ascribed to a lack of motivation, few incentives and poor access to registration services. Low registration of cancer cases among women and girls mirrors this pattern: studies have found a disproportionately higher registration of cancer cases in boys than girls.

Moreover, cancer mortality is harder to track down in rural areas without access to health care or medical certification of deaths, leading to an under-reporting of cases. It is also driven in part due to a social stigma around chronic conditions like cancer, which risks abandonment and ostracisation for vulnerable communities, including women. In new registries that opened in Punjab’s Sangrur and Chandigarh, the staff had to tiptoe around the ‘C word’ and phrase more vague questions about treatments. One personnel told the 2020 researchers, “...it’s a very dreaded thing...sometimes even the relatives don’t want it to be disclosed to the patient. They won’t take the name of cancer...they ask us not to tell the patient.”