Finding Cancer at Home

Supplements and Featured Publications, Special Issue: Health Information Technology — Guest Editors: Sachin H. Jain, MD, MBA; and David B, Volume 16,

One medical student’s search for answers about her sister’s cancer leads her to realize that health information exchange can support the development of more sophisticated approaches to identifying root causes of disease.

In March 2010 I achieved a lifelong dream and was accepted to medical school. My excitement was tempered when weeks later my sister, Jaya, was diagnosed with Hodgkin Lymphoma. Jaya was finishing her third year of nursing school when she noticed a small lump above her collarbone that changed my family’s life forever.

The diagnosis came as a shock. It was hard to reconcile the fact that my 21-year-old younger sister—who had just run her first half-marathon, consistently ate healthily and insisted on sleeping 8 hours every night, and who felt well—had a hulking mass in her chest and cancerous lymph nodes throughout her neck. Overnight, she went from being the picture of health to a cancer patient. In hindsight we realized that she had been experiencing insidious symptoms for the preceding 2 years—a rash that was attributed to eczema, sweats that were chalked up to anxiety, and chest pain that was dismissed as “spasms”—but the possibility of cancer had never entered our equation.

We learned from Jaya’s oncologist and our own research that Hodgkin Lymphoma was one of the “better’’ cancers to have. She had the most common subtype, and her prognosis was good (although we were never sure exactly how good, as she told her doctor she did not want to hear the numbers). Regardless of how far the cancer had spread, she would start with 12 infusions of chemotherapy over 6 months.

When we or our loved ones receive the news of a difficult diagnosis, it is only natural to question why, how, and from where this condition arose. No one in our family had lymphoma. I thought it was a strange coincidence that my good friend, who lived only a few streets away from us for most of our lives, was diagnosed with Hodgkin Lymphoma only 2 years before my sister. When I started talking to people from our small hometown of Hampstead, New Hampshire, population 9000, I discovered a handful of other young people who had been diagnosed with lymphoma in recent years. Adding to our suspicions, several of our neighbors had been diagnosed with cancer—one man died of stomach cancer, a husband and wife both developed breast cancer, and 2 dogs were diagnosed with some type of cancer—all within a few years’ time. I set out to learn more about cancer clusters. When I began my research I knew nothing about what constituted a “cluster” other than a greater number of cancer cases than would be expected in an area. I began pondering basic questions. “How many people constitute a cluster?” “To whom do you report a case?” And “What environmental factors might be implicated?” Jaya was a continuous source of motivation as I pursued these questions. While she had much else on which to focus during this time—school, work, and studying for her nursing board exams—she always encouraged me and was eager for updates on my progress. My search was humbling, at times discouraging, and, in the end, eye-opening.

The first place I began asking questions was the oncologist’s office. I asked Jaya’s doctor if he knew anything about cluster investigations; if he ever noticed patterns of patients coming from certain towns or neighborhoods; if there were any known environmental pathogens that could cause lymphoma and other types of cancer. His answer did not satisfy: “Cancer is common and identifying the root cause of a specific patient’s cancer is most often impossible.” I asked if his office kept an electronic record of patients’ zip codes with their diagnoses, and he explained that he “thought so” but couldn’t release that to me because it would violate patient privacy. My sister’s chemotherapy infusion nurse explained to me that “testing your water is very expensive” and therefore not a very practical option.

My next step was to make some phone calls to the local health department, but I wasn’t sure which health department to call. Between 1988 and 1996 we lived in Atkinson, New Hampshire; between 1996 and 2008, we lived in Hampstead, New Hampshire; and in 2008, my parents moved to Hampton while we were away at college. I called the town of Hampstead first, and they referred me to the Centers for Disease Control (CDC). The CDC referred me to the NH Department of Health and Human Services (DHHS). While waiting for my phone calls to be returned, I began researching my questions online.

One of my most startling discoveries was finding out about the cancer cluster investigation that had taken place 8 years prior on Providence Hill Road in Atkinson—the street on which we lived between 1988 and 1996. The town tax collector had compiled a list of 32 people within a half-mile radius of our house who had been diagnosed with some form of cancer since 1980 (including a man who lived across the street from us who died of lymphoma in 2002). While the state determined that it was not a true cluster, the investigation brought about enough public attention and awareness for several residents to have their private wells tested. Those tests revealed radon levels 15 times higher than a proposed federal standard and 30 times the level at which the state of New Hampshire advises that residents be concerned. Test results from the public water system disclosed levels of radon 20 times the level of state concern. While the EPA recognizes that high levels of radon pose a risk for cancer (mainly stomach and lung), they have yet to finalize a proposed drinking water standard that would compel public water suppliers to install equipment to reduce radon levels.

Around the corner from our house in Atkinson was an old dump site that was a cause of concern for people in the area, especially when taking into account the volume of local cancer diagnoses. Some suspected that Massachusetts-base companies were illegally dumping hazardous waste there in the 1980s. I spoke with a geologist specializing in hazardous waste site remediation, who told me that the DHHS did not have the resources or authority to investigate old dump sites without clear proof of illegal activity or contamination nearby.

In light of these discoveries, I was left wondering how many people it would have taken for our neighborhood in Atkinson to be considered a true cluster, and how many people—like us—might have moved and developed cancer afterward. A state epidemiologist explained that there were 3 considerations when investigating a potential cluster: if the same type of cancer is occurring across patients, if it is a rare cancer, and if it is being diagnosed in an unusual age group. I was relieved to hear about the state cancer registry, which collects data from every hospital in the state to analyze cancer diagnoses in different geographic areas. However, learning about the limitations of cancer epidemiological research was anything but reassuring.

There is a 2-year time lag on the data that can be analyzed from the cancer registry. This lag can be attributed in part to the fact that many hospitals do not have electronic records, and sifting through paper records is extremely labor-intensive. Data is not collected on a person’s previous residences, so individuals who move are not considered in investigations. Even though my sister had barely spent any time in Hampton, she will go into the registry as a Hampton resident since that was her address at the time of diagnosis. In the small towns in which we lived, even 1 or 2 more cases of a particular cancer in a given year would be statistically significant—but it would take a lot more than that for DHHS to raise its eyebrows and start an investigation, as it would be resource-intensive and fruitless to chase down the cause of cases that were likely due to chance.

Cancer epidemiology is complex, and I understand that resources are limited. However, I was confused when I learned that the index of suspicion during a cluster investigation is only raised when the same specific type of cancer is being seen across many different people. When I told our state cancer epidemiologist about all of the cancer on our street, alarm bells did not go off because most people had different types. Couldn’t 1 environmental pathogen cause different types of cancer? Radon can cause lung and stomach cancer. Arsenic is linked to skin, bladder, and lung cancer. It seems that patterns of certain cancers would be important to look at from an epidemiological standpoint.

The DHHS cancer epidemiologist was kind enough to prepare a report for me that included information on incidence rates of every type of cancer in our 3 hometowns—Atkinson, Hampstead, and Hampton—from 1995 to 2007. These reports compared the town incidence rates to the state incidence rates. Surprisingly, there was an elevated incidence of Hodgkin Lymphoma in Hampton, the town to which my parents moved in 2007. The standard incidence ratio (SIR) was 2.14, indicating that over twice the number of cases were found than expected (this translated to 6 extra cases). However, if my sister’s cancer was due in part to an environmental exposure, it could not have been in Hampton, as she had not lived there long enough for the cancer to have developed. The other towns we lived in had higher rates of other types of cancer, but not Hodgkin Lymphoma—at least until 2007. Data from more recent years was not yet available in the registry.

My search left me with more questions than answers. I wanted to see the incidence rates of cancer from the past 2 ½ years in the towns in which we lived. I wanted to be able to look at things at a more local level—what if all of the cancer cases in a town were from the same 1 or 2 neighborhoods? Perhaps aggregated town-level data obscured relevant patterns. I wanted to know more about the relationship between environmental exposures and Hodgkin lymphoma, a research topic on which we have barely scratched the surface. I wanted to better understand why there was a lag in translating research that has been done on known carcinogens into rules and regulations that could protect us.

My search did not lead to the discovery of a cancer cluster. It did not unearth a carcinogenic contaminant. I did not catch any bandits red-handed as they tried to dump hazardous waste on our country roads. My sister recently finished her chemotherapy treatments and we are ready to put this small nightmare behind us. Maybe Jaya was just a victim of bad luck—we may never know. What I did take from my experience are lessons that have encouraged me to be a more proactive citizen and will hopefully make me a better physician.

There are a number of disincentives to pursuing the “how” and “why” questions that plague us when we have personal or professional encounters with illness: the fear of looking foolish or ignorant; the decline in local property values if an environmental exposure is identified; the substantial resources expended during an investigation; the time and effort that goesinto a search that may end in futility. However, we should persist in our curiosity about the sources of diseases that afflict us.

Rather than resigning ourselves to the complexities with which we are faced, we should work to resolve them. I am now a strong advocate for the use of electronic health records and electronic health information exchange, as I have seen the implications they may have for faster and more sophisticated analyses of cancer registry data. I now encourage people I know with private wells to have their water checked regularly. When people tell me about neighborhoods where there “seems to be a lot of cancer,” I urge them to contact DHHS.

In time, we may find the answers we are looking for—if we continue to pursue our questions.

Acknowledgment

The author would like to thank Peter McGlew for his investigative work and expertise he lent on the topic of hazardous waste site remediation and environmental contaminants.

Author Affiliation: From Tufts University School of Medicine, Boston, MA.

Funding Source: None reported.

Author Disclosure: The author reports no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

Authorship Information: Concept and design; acquisition of data; analysis and interpretation of data; drafting of the manuscript; critical revision of the manuscript for important intellectual content.

Address correspondence to: Katlyn L. Nemani, BA, 145 Harrison Ave., Boston, MA 02111. E-mail: katlyn.nemani@tufts.edu.