Oswego Foodborne Outbreak, 1940

Analysis of Gastrointestinal Illness Following a Church Supper

Author

Affiliation

Core-Data

Abstract

A foodborne illness outbreak occurred among attendees of a church supper in Lycoming, Oswego County, New York, in April 1940. A retrospective cohort study was conducted among 75 of 80 attendees to collect demographic information, food consumption histories, and illness status. Gastrointestinal illness, including nausea, vomiting, and diarrhea, was reported by 46 participants (61%). Food-specific attack rates were calculated, and vanilla ice cream was associated with the highest attack rate (79.6% vs 14.3% among non-consumers). The epidemic curve indicated a point-source exposure with a median incubation period of 4 hours. Findings are consistent with Staphylococcus aureus food poisoning. This study highlights the importance of food safety practices in preventing outbreaks at community events.

Keywords

Epidemiology, Foodborne Illness, Oswego Outbreak

Introduction

Foodborne outbreaks are a significant public health concern, capable of causing widespread gastrointestinal illness and placing substantial demands on local health services. Prompt recognition and systematic investigation of such outbreaks are essential to identify the source of infection and implement measures to prevent further cases.

On 19 April 1940, the local health officer in Lycoming, Oswego County, New York, reported an outbreak of acute gastrointestinal illness following a church supper held the previous evening. Preliminary information indicated that all reported cases had attended the supper, while non-attendees remained unaffected. The supper was held in the church basement, with food contributed by multiple members of the congregation and served over several hours.

A rapid epidemiological investigation was initiated to determine the likely source of the outbreak, describe the characteristics of illness, and provide evidence to guide public health action.

Objective: The purpose of this study was to identify the probable food vehicle and describe the epidemiological features of the outbreak to inform preventive and control measures.

Methods

Study Design

A retrospective cohort study was conducted among individuals who attended the church supper in Oswego. Data were collected through structured interviews to assess food exposures and illness outcomes following the event.

Case Definition

A case was defined as any person who attended the church supper held in the basement of the village church in Lycoming, Oswego County, New York on 18 April 1940, and developed acute gastrointestinal symptoms, including nausea, vomiting, diarrhoea, or abdominal cramps, within 48 hours after the event.

Data Collection

Information collected included demographic characteristics (age and sex), time of meal consumption, specific food items eaten, illness status, and time of symptom onset.

Statistical Analysis

Descriptive statistics were used to summarise demographic characteristics of the study population. Food-specific attack rates, risk ratios, and corresponding 95% confidence intervals were calculated to assess associations between food exposures and illness. All analyses were conducted using R version 4.5.

Results

Summary of Participant Characteristics and Demographics

Characteristic	Overall N = 75¹	Ill N = 46¹	Not Ill N = 29¹	p-value²
Outbreak of Gastrointestinal Illness in Oswego County, 1940
Age (In Years)	36 (16, 58)	39 (17, 59)	35 (14, 50)	0.3
Sex of the Participant				0.15
Female	44 (59%)	30 (65%)	14 (48%)
Male	31 (41%)	16 (35%)	15 (52%)
Incubation Period (In Hours)	4.0 (3.0, 6.0)	4.0 (3.0, 6.0)	NA (NA, NA)
Source: Oswego An Outbreak of Gastrointestinal Illness Following a Church Supper (updated 2003): https://www.cdc.gov/eis/casestudies/xoswego.401-303.student.pdf
¹ Median (Q1, Q3); n (%)
² Wilcoxon rank sum test; Pearson’s Chi-squared test; NA

Of the 80 individuals who attended the church supper, 75 (94%) were interviewed. Among these, 46 (61%) reported gastrointestinal illness. The median age of participants was 36 years (IQR: 16–58); median age among ill participants was 39 years (IQR: 17–59) and among non-ill participants 35 years (IQR: 14–50) (p = 0.30). Females comprised 59% of participants, and males 41%, with no significant difference in illness status by sex (p = 0.15).

Epidemic Curve

The epidemic curve showed that cases occurred within a short time period following the supper. The median incubation period among cases was 4.0 hours (IQR: 3.0–6.0).

Food-Specific Attack Rates

Food-specific Attack Rate Table
Oswego County Outbreak Investigation, 1940
Food Item	Ate Food				Did Not Eat Food				Attack Rate Ratio
Food Item	Ill	Not Ill	Total	Attack Rate (%)	Ill	Not Ill	Total	Attack Rate (%)	Attack Rate Ratio
Vanilla Ice Cream	43	11	54	79.6	3	18	21	14.3	5.6
Cakes	27	13	40	67.5	19	16	35	54.3	1.2
Jello	16	7	23	69.6	30	22	52	57.7	1.2
Brown Bread	18	9	27	66.7	28	20	48	58.3	1.1
Fruit Salad	4	2	6	66.7	42	27	69	60.9	1.1
Baked Ham	29	17	46	63.0	17	12	29	58.6	1.1
Cabbage Salad	18	10	28	64.3	28	19	47	59.6	1.1
Coffee	19	12	31	61.3	27	17	44	61.4	1.0
Mashed Potato	23	14	37	62.2	23	14	37	62.2	1.0
Spinach	26	17	43	60.5	20	12	32	62.5	1.0
Rolls	21	16	37	56.8	25	13	38	65.8	0.9
Water	13	11	24	54.2	33	18	51	64.7	0.8
Milk	2	2	4	50.0	44	27	71	62.0	0.8
Chocolate Ice Cream	25	22	47	53.2	20	7	27	74.1	0.7
AR = Attack Rate; Exposure = Ate the food (Y/N)

Attack rates for each food item were calculated. Vanilla ice cream had the highest attack rate among consumers (79.6%) compared with non-consumers (14.3%), resulting in an attack rate ratio of 5.6. Other food items had attack rates ranging from 50% to 69%, with attack rate ratios close to 1.0.

Sex and Food-Specific Attack Rates

Stratified Food-specific Attack Rates by Sex
Oswego-style Outbreak Analysis
Food Item	Sex	Ate Food				Did Not Eat Food				Attack Rate Ratio
Food Item	Sex	Ill	Not Ill	Total	Attack Rate (%)	Ill	Not Ill	Total	Attack Rate (%)	Attack Rate Ratio
Baked Ham	Female	19	7	26	73.1	11	7	18	61.1	1.2
Baked Ham	Male	10	10	20	50.0	6	5	11	54.5	0.9
Brown Bread	Female	9	4	13	69.2	21	10	31	67.7	1.0
Brown Bread	Male	9	5	14	64.3	7	10	17	41.2	1.6
Cabbage Salad	Female	13	3	16	81.2	17	11	28	60.7	1.3
Cabbage Salad	Male	5	7	12	41.7	11	8	19	57.9	0.7
Cakes	Female	19	5	24	79.2	11	9	20	55.0	1.4
Cakes	Male	8	8	16	50.0	8	7	15	53.3	0.9
Chocolate Ice Cream	Female	19	11	30	63.3	11	3	14	78.6	0.8
Chocolate Ice Cream	Male	6	11	17	35.3	9	4	13	69.2	0.5
Coffee	Female	12	6	18	66.7	18	8	26	69.2	1.0
Coffee	Male	7	6	13	53.8	9	9	18	50.0	1.1
Fruit Salad	Female	1	1	2	50.0	29	13	42	69.0	0.7
Fruit Salad	Male	3	1	4	75.0	13	14	27	48.1	1.6
Jello	Female	11	3	14	78.6	19	11	30	63.3	1.2
Jello	Male	5	4	9	55.6	11	11	22	50.0	1.1
Mashed Potato	Female	13	7	20	65.0	17	7	24	70.8	0.9
Mashed Potato	Male	10	7	17	58.8	6	7	13	46.2	1.3
Milk	Female	1	0	1	100.0	29	14	43	67.4	1.5
Milk	Male	1	2	3	33.3	15	13	28	53.6	0.6
Rolls	Female	13	6	19	68.4	17	8	25	68.0	1.0
Rolls	Male	8	10	18	44.4	8	5	13	61.5	0.7
Spinach	Female	18	8	26	69.2	12	6	18	66.7	1.0
Spinach	Male	8	9	17	47.1	8	6	14	57.1	0.8
Vanilla Ice Cream	Female	27	4	31	87.1	3	10	13	23.1	3.8
Vanilla Ice Cream	Male	16	7	23	69.6	0	8	8	0.0	Inf
Water	Female	8	4	12	66.7	22	10	32	68.8	1.0
Water	Male	5	7	12	41.7	11	8	19	57.9	0.7
AR = Attack Rate; Exposure = Ate the food (Y/N)

Stratified analysis showed higher attack rates among both females and males who consumed vanilla ice cream. Attack rates for other foods showed no substantial differences by sex.

Logistic Regression Model

Discussion

This outbreak investigation identified a strong association between gastrointestinal illness and consumption of vanilla ice cream at a church supper in Oswego County. Individuals who consumed vanilla ice cream experienced a substantially higher attack rate than those who did not, and this pattern was consistent across sex strata. The epidemic curve suggested a point-source exposure, indicating a common vehicle of transmission.

The short incubation period, predominance of gastrointestinal symptoms without fever, and rapid recovery are consistent with foodborne intoxication, most likely caused by Staphylococcus aureus enterotoxin. Although laboratory confirmation was unavailable, the epidemiological findings and clinical presentation support this interpretation. Dairy-based foods are recognised vehicles for staphylococcal toxin transmission, particularly when temperature control is inadequate.

Strengths of the investigation included a clearly defined exposed population, high participation rate, and detailed food consumption data. Limitations included the retrospective study design, potential recall bias, absence of microbiological testing, and limited sample size, which restricted multivariable analysis.

These findings underscore the importance of proper food handling, temperature control, and hygiene during community events to prevent similar outbreaks.

Conclusions

This investigation identified vanilla ice cream as the most likely vehicle of a foodborne outbreak among church supper attendees in Oswego County in 1940. The epidemiological and clinical features were consistent with staphylococcal food poisoning. These findings reinforce the importance of proper food handling and temperature control during community events.