House on Fire (18 page)

I wish you a good meeting, a good campaign and I look forward to the day when we all can enjoy a smallpox-free Bihar State.
³

On the morning of October 15, workers departed from the headquarters of 1,462 PHCs. Each PHC searched twenty to twenty-five villages a day for six days, from Monday through Saturday. By the end of the week, over two hundred thousand villages had been visited in the two states, and the PHCs started to tally the results. Within days the results began to flow into the seventy-one districts. After district totals were compiled, the results were forwarded to the states. It was like a river drainage basin with hundreds of thousands of small streams forming creeks of information, the creeks forming rivers; the delta they flowed into was the assembly of all of this information for these two states.

In Uttar Pradesh, a state meeting of field personnel was scheduled for November 5. The New Delhi staff began to worry that the meeting might be chaotic if we received the district reports from this first search at the meeting itself. The caliber of revisions in methodology for the second search might be compromised for lack of time to properly consider the results. It was decided that during the week immediately following the search, I would canvas as many districts in Uttar Pradesh as possible in five days. Then we would review what I found and use this limited sample as the basis for preliminary recommendations for the second search.

The program provided me with an Indian-made jeep and a driver. The driver spoke no English, and I soon discovered he was unable to read road signs in either English or Hindi. The Indian roads were a continuing challenge: a mixture of people walking, people on bicycles, rickshaws, scooters at times carrying four or five people, motorcycles, farm vehicles, bullock carts, large trucks, buses, cattle, camels, goats, chickens, varying widths of roadway, potholes, disregard for road rules—in short, chaos. This would have been trying even if we were not also contending
with heat, dust, and diesel fumes, a cacophony of noise, and constant changes in speed. (We were stopped at one roadside safety check where the only test was whether the horn worked.)

At each district headquarters we would get detailed instructions for the driver on how to get to the next district. Yet getting lost several times a day became the constant. The variable was the length of time that elapsed before we knew we were lost. In scenes that reminded me of a Laurel and Hardy movie, we would stop, I would ask for directions in English, and the person would advise the driver on how to proceed in Hindi. For five days and nights, I gathered material and reviewed the findings as we drove. We stopped for meals, but to save time we did not bother to stop to find sleeping facilities. Rather, one of us would try to sleep while the other drove. Arriving at a new district headquarters, we would sleep in the jeep until the offices opened. After I had met with the district officers, we would be off to the next district.

It was in the middle of one of these night drives, with no moonlight, when my height presented a problem that could have been fatal. I was driving, my knees pressed against the dashboard, and one knee nudged the light switch, turning out the headlights; in the darkness, I was not immediately aware of why the lights had gone out. There was a moment of panic at forty-plus miles per hour, and then I realized what had happened and turned the lights back on. Despite the narrowness of the two-lane paved road, which barely allowed two trucks to pass each other, we were still on it! I have no idea how we escaped hitting the trees that the British had so thoughtfully planted long ago, which lined each side of the road.

The results of our trip were at first intriguing, then sobering, and finally scary. The search teams were finding far more smallpox cases than we had anticipated. We had based some of our expectations and many of our predictions on the passive reporting of the previous weeks. The number of new cases reported each week, for Uttar Pradesh and Bihar combined, had been in the hundreds. Even if the true figure was twice that number, it would not exceed the capacity of the containment teams already trained.

It was not simply that outbreaks were larger than projected, or that
areas with smallpox had more outbreaks than anticipated—both of which turned out to be true. The real surprise was that new outbreaks were being discovered in PHCs that did not even know they had smallpox and thought they had been free of the disease for some time. This was the flaw of the passive reporting system. Health workers at the PHCs were waiting for patients to come to them. But smallpox patients were not showing up at the health centers because they knew that the health service could not help them.

I worked on the figures as we traveled, trying to understand what they meant in terms of containment efforts but also wondering what we would tell searchers for the next month's search. Would there be more surprises, or had this six-day search provided us with the truth? The last day of the five-day circuit required a stop in the city of Agra to meet with district health officers to discuss search results. We did not even have the time to drive past the Taj Mahal.

As the driver began the drive back to New Delhi, I started to analyze the figures just received. We had not even made it out of town when fatigue overtook the driver and he ran into a light pole, creasing the radiator and making the vehicle inoperable. While we were shaken by the incident, seat belts prevented serious injury. With a focus on the work at hand that is difficult to comprehend thirty years later, the driver and I shrugged off the incident and immediately shifted to solving the problem. Within twenty minutes, I had secured a district car and driver, and continued on my own to New Delhi. The driver stayed behind to fill out the accident report and arrange for repairs on the vehicle.

After I reached New Delhi, I began to feel the effects of the five-day trip. I developed herpes zoster (shingles) in a nerve pattern that involved the right side of my face below my eye. Within days I proceeded to Lucknow, the capital of Uttar Pradesh, to attend the first state meeting. The zoster lesion was weepy and hurt a great deal. At the Lucknow meeting, I was glad to encounter Dr. Gordon Meiklejohn, chairman of the Department of Medicine at the University of Colorado and one of the world's experts on herpes viruses. How could I have been so fortunate that a world-class herpes expert had volunteered as a short-term smallpox worker in India? He examined me and gave me reassurance. Months
later he told me that he had nearly sent me back to the States because he feared the virus might extend into my eye.

The sample drawn from the five-day trip had been a warning. But at the meeting, as the reports from the fifty-four districts came in, we realized that we were in fact facing a disaster. In September, via the existing reporting system, Uttar Pradesh had reported 437 cases of smallpox. Now, just one month later, searchers had found 5,989 new cases. To put this in perspective: in only two of the previous ten years had Uttar Pradesh reported more than 7,000 cases in an entire year. Now, to the surprise and regret of the smallpox workers, it was found that 87 percent of the reporting districts—including 1,483 villages and 42 municipalities—had been harboring smallpox.

At Bihar's state meeting, the search team reported 3,826 new cases. However, the figures were incomplete because of inadequate staffing, so it was not even clear how many of Bihar's 587 PHCs had smallpox cases. Even with incomplete reporting, the search revealed that 477 of the 50,000 villages canvassed in Bihar, and 13 of its 103 municipalities, had smallpox cases. Smallpox was reported in sixteen of Bihar's seventeen districts.

The first search had identified almost ten thousand new cases in only two states. In the two states combined, 90 percent of all districts had smallpox and two thousand villages were involved, some with multiple outbreaks. In addition, 10 to 26 percent of all urban areas had smallpox. At a time when we anticipated a low point in numbers, we found smallpox everywhere.

On the one hand, we were euphoric at having pulled off something this unprecedented. Everyone praised the searchers for finding new cases, and they in turn were pleased with their results. In Uttar Pradesh in particular, health workers seeking smallpox had managed to get into almost 99 percent of the state's 140,000 villages and do at least a cursory search for smallpox in the short span of six days. That was no small feat.

On the other hand, the numbers signaled a daunting amount of work to respond to the search findings. We were in well over our heads. Indeed, the program did not have enough containment teams to travel to every new outbreak, even if they did no vaccinations.

Figure 11.
Smallpox reports from weeks 34 to 47 in Uttar Pradesh and Bihar, India, 1973

Some smallpox workers at the state and federal levels argued that we should cancel the next month's search and concentrate on containment, since it made no sense to find more cases if we could not control them. Some even suggested that we stop this form of surveillance entirely, as it made the medical system look inadequate for not reporting the cases earlier. However, the situation was overwhelming precisely because this was the best surveillance effort India had ever undertaken. The search had revealed what actually existed, not what we hoped existed. Keeping the truth in front of us would force us to respond appropriately. Our response should be to improve containment, not dumb down surveillance. The lesson for combating smallpox, and indeed for all public health programs, is that you can't form an effective response until you know the truth.

The central smallpox team decided to put at least a cursory effort into containment after the first search. We did our best to increase the
number of containment teams, and they were told to concentrate only on the families with smallpox and the adjacent houses, rather than trying to vaccinate twenty to thirty nearby households, as the guidelines specified.

At the same time we tried to draw lessons from the first search in order to improve the next one, which was to take place just a week after the meeting. For example, searchers had noticed that schoolchildren were some of the best informants. They knew what was happening in their own neighborhoods and were not as reticent as their parents. There was also a hint of competition in the classroom to be the person giving the information. One new directive was to ask children if they knew anyone who had smallpox.

Everyone avoided discussing the possibility that the problem was so immense that it might be unmanageable. Instead, the focus was on how to address the barriers to eradication.

Many years later, Harlan Cleveland, a respected American political scientist, diplomat, educator, and author, observed that global health workers were fueled by “unwarranted optimism.” That phrase well describes this band of smallpox workers as they took on the biggest public health challenge they had ever faced. It wasn't a case of putting a spin on the reality of the situation in order to fool others. If anything, we were fooling ourselves—though not totally. After the first search, the SEARO office in New Delhi sent a request to the WHO office in Geneva to provide up to sixty additional special epidemiologists. Needless to say, when the Geneva officials received the request, they were shocked.

Armed with new information and with the experience of the first search fresh in mind, the search teams in Uttar Pradesh and Bihar moved out across the landscape for the second search on November 12. The results showed that the search teams had already improved their skills. Reporting was more complete. In Uttar Pradesh, again almost 99 percent of the villages were searched, but the searches were more effective.

During the three weeks between searches, another 800 new cases had been reported through the passive reporting system that was already in place. Hopefully this number constituted most of the backlog, and the numbers were now close to accurate. In fact, the second search revealed another 1,711 new cases in Uttar Pradesh. The situation was even worse in Bihar, where 2,459 new cases were found. Some of the new cases were within outbreaks found the month before. Others were in new villages but were quickly traced to outbreaks uncovered the previous month that had been poorly contained. However, far too many of the new cases had no connection to recent outbreaks. This meant that these cases had either been missed the month before or had resulted from outbreaks that had been missed.

The program was still far short of the ability to respond adequately to the level of smallpox being found. Resources for containment were stretched to the limit, and health personnel, supervisors, and trainers were not available at the levels required. Should the containment teams stay at an outbreak until all susceptible persons had been vaccinated, even if it meant not responding to all outbreaks? Or was their time best spent going to all outbreaks but doing an inadequate job at each? Either decision was an unprofessional, slap-dash approach to a very serious situation.

The compromise, unfortunately, was to do both, that is, to do an inadequate job at every outbreak visited and not reach some outbreaks at all. The smallpox team's “unwarranted optimism” allowed us to hope that vaccinating even a small number of people who were at highest risk would impact disease transmission, just as small increases in humidity and reductions in wind velocity can dramatically improve the efficiency of firefighting efforts. On this point, our optimism was trumped by reality.