Exposure to cold weather during a mass gathering in the Philippines

Abstract

Problem

The visit of Pope Francis to the Philippines in January 2015 coincided with a tropical storm. For security reasons, the only road in and out of the area was closed 14.5 hours before the Pope’s arrival. This meant that people had to wait for many hours with little shelter at the site. Medical teams in the field reported high numbers of people with cold stress during the mass gathering.

Approach

To review the event from a public health perspective, we examined the consultations made by medical teams in the field and interviewed key stakeholders, focusing on cold stress as a public health risk.

Local setting

The key reason for the Pope’s visit to Palo and Tacloban was the devastation caused in these cities by typhoon Haiyan in 2013. We estimated that the visit attracted 300 000 people. The medical teams were advised to consider cold stress risks two days before the event but no other measures were taken.

Relevant changes

Of the 1051 people seeking medical care, 231 people were experiencing symptoms of cold stress. People with cold stress ranged from 2 to 89 years of age and were more likely to be female than male, 173 (75%) versus 57 (25%).

Lessons learnt

Planning for mass gatherings should consider a wide range of public health risks, including cold stress. Improved data collection from the field is necessary to maximize the benefits of post-event evaluations and improve public health preparedness. Security measures to ensure the safety of key figures must be balanced with public health risks.

Introduction

Between January 15 and 19, 2015, Pope Francis visited the Philippines. On January 17 the Pope visited the cities of Palo and Tacloban, which coincided with the landfall of tropical storm Mekkhala. During his visit in Palo and Tacloban, he conducted an open-air service, visited the archbishop’s residence and Palo Cathedral. The visit attracted large crowds and specific planning was done ahead of time. The pre-event planning focused on hyperthermia, crowd safety and control, but had not considered the risk posed by cold weather. Due to security procedures, the only road in and out of the site was closed to vehicles 14.5 hours before the Pope’s arrival at 18:00 hours, effectively trapping crowds in the area. Before the road closure, attendees were sent by the event organizers to a designated drop-off point 2 km from the site; resulting in many people walking to the site and waiting for the service in the rain for around 19 hours. By 18:00 hours the evening before the visit, all 56 medical teams in the field were prepositioned evenly in the area. The teams used paper logs to record patient consultations. The Philippines Department of Health advised medical teams to consider cold stress risks two days before the event but no additional measures were taken.

Approach

To assess the number of people seeking care for cold stress during the event, we reviewed the patient logs. Data were analysed separately by the log field “diagnosis” and then by “chief complaint and/or symptom”. The latter was used as a proxy for formal diagnosis. Logs where both fields were blank were excluded. Analysis was conducted using Stata version 12.1 (StataCorp. Lp, College Station, United States of America).

Cold stress was identified when either a medical doctor recorded a diagnosis of cold stress or the log sheet described signs and symptoms consistent with cold stress and no other clinical diagnosis was apparent. Cold stress was diagnosed when axillary temperatures below 36.0 °C were recorded on presentation. Monitoring of vital signs could not be linked to specific log sheets and there were no available data on recovery times.

Twenty-four semi-structured and informal interviews were conducted with key informants to validate findings from log sheets. Respondents included focal points from the Philippines Department of Health, field medical team leaders, referral hospitals, police, the Archdiocese of Palo and attendees.

We estimated the total crowds in Tacloban and Palo at 300 000 people. Pre-event planning had expected one million attendees. Crowds were estimated in four key areas at: (i) the site of the service (designed for 160 000 individuals but officials in attendance suggest crowds were closer to 200 000 people); (ii) the archbishop’s official residence (capacity 3000); (iii) Palo Cathedral (capacity 1500); and (iv) the route between each stop. For the people lining the road, we assumed four people per square metre.¹ The route taken by the Pope was approximately 11 km long and, based on the estimates from medical teams along the route, approximately 80% of one side of the road was lined with bystanders approximately three metres deep. We thus estimated that 105 600 people were present along the route.

We searched social media (i.e. Facebook and Twitter) for public advisories related to the visit. Almost all suggested that people wear light and comfortable clothing and warned of hyperthermia. Umbrellas and opaque backpacks were prohibited from the site.

Local setting

Tacloban is situated approximately 360 km south-east of Manila and Palo is a further 12 km south. Typhoon Haiyan devastated these communities in early November 2013 and was the key reason for the Pope’s visit.² Two days before the visit, it was forecast that tropical storm Mekkhala would make landfall about 100 km north-east of Tacloban airport.³

The weather station located at the airport monitored ambient temperature, which remained above 22 °C from 03:24 hours on 16 January to 08:19 hours on 17 January. Heavy rain was recorded between 16:00 and 18:00 hours, on 16 January, and was followed by continuous drizzle until 04:00 hours.⁴

Relevant changes

About half of the medical teams submitted logs (31/56; 55%). These teams were spread evenly throughout the crowd. There were 1051 recorded consultations and the log field “chief complaint and/or symptom” was more often complete than for the field “diagnosis” (1006/1051; 96% versus 526/1051; 50%, respectively). A smaller subset of logs mentioned cold stress even when the “diagnosis” field was blank (38/526; 7%).

Most consultations were at the site of the service and done by staff from the four field hospitals (744; 71%), followed by medical stations (258; 24%) and ambulances (49; 5%) along the route, in the cathedral and at the archbishop’s residence. Cold stress accounted for 22% of consultations (231). No cases were referred for further care and there were no deaths.

People with cold stress ranged from 2 to 89 years of age and were more likely to be female than male (173; 75% versus 57; 25%). The highest proportion was seen among women older than 60 years (Table 1).

Table 1. People with cold stress during a mass gathering event in the Philippines, January 2015.

Characteristic	Estimated population at mass gathering			No. of cases			Cases per 100 000 population
	Female	Male		Female	Male		Female	Male
Age group, years
0–19	60 507	63 144		81	34		133.87	53.85
20–59	76 482	77 016		45	17		58.84	22.07
≥ 60	12 282	10 569		39	6		317.54	56.77
Not recordeda	–	–		8	0		–	–
Total	149 271	150 729		173	57		–	–

^a There were nine cold stress cases with unrecorded age (eight females, one unrecorded sex).

We learnt from interviews that those present sought medical attention only when they felt unable to continue tolerating the cold and that they had been regularly advised to protect themselves from the weather. Most patients needed assistance to reach the health station and were wet and wearing light clothing on arrival. Blankets and hot drinks were provided on arrival (before recording body temperature). Personal medical histories were not recorded. While the first cold stress case outside the site was reported at 22:00 hours (four hours after the road closure), at the field hospitals where most cases were seen, the first case was at 04:15 hours (almost 12 hours after the road closed). Cases continued to be reported throughout the early morning and peaked after the Pope departed at 13:00 hours (Fig. 1).

Fig. 1.

People with cold stress and key events during a mass gathering in the Philippines, January 2015

Note: Three weather bulletins (issued at 05:00, 11:00 and 17:00 hours on 17 January) reported rainfall and wind speed: 7.5 –20 mm/h; 100 kph with gusts to 130 kph respectively.

Cold stress affected people attending the Pope’s visit in Tacloban and Palo. The main lessons learnt from this event are summarized in Box 1. Most cases were seen in field hospitals at the site of the open air service, which was where people were exposed to the wind and rain for the longest period. While we found the proportion of people with cold stress increased with age, consistent with related studies,^5,6 we saw the highest absolute number of cases among young people, which is consistent with the Philippine’s demographic profile (41% of the population are younger than 20 years).⁷ Hypothermia has been associated with poor nutrition and female sex.⁵ Poor nutrition was common in Tacloban and Palo as a result of typhoon Haiyan. About 5.6 million survivors of all ages were at risk of food insecurity.^8,9

Box 1. Summary of main lessons learnt.

• Even in tropical countries, pre-event planning for mass gathering events must consider a wide range of public health risks, including cold stress.

• Improved data collection in the field is necessary to maximize the benefits of a post-event evaluation and improve public health preparedness for mass gathering events.

• Security measures must be balanced with public health risks.

Our study is likely to underestimate the number of people with cold stress. We received information from only half of the medical teams but know the other teams provided medical care to people. The data received were representative of teams at all locations and given the conditions it is likely those not reporting saw people with similar conditions. We were unable to find previous reports of so many people with cold stress associated with a mass gathering in a tropical country.

Lessons learnt

This experience has provided important lessons for the Philippines. Tropical cyclones, which commonly result in large-scale evacuations to high density shelters, are permanent hazards. Our study demonstrated cold stress is a legitimate public health concern.

Following the gathering, local health authorities recognized the need to broaden the range of public health risks in pre-event planning. Cold stress was not considered a risk despite the visit occurring towards the end of the cyclone season. The Philippines Department of Health recognized the importance of improving clinical data collection to respond more effectively to a changing situation. Provision of standard individual patient information sheets in addition to summary logs is now being considered for similar events. We recommended procuring supplies (such as thermal blankets) for the clinical management of cold stress cases as a general preparedness measure.

Security measures, such as closing roads and banning umbrellas, increased people’s exposure to the weather. Public health risks need to be considered when planning security measures for mass gatherings.