Increased consumer demand for convenience and ready-to-eat food, along with changes to consumer food purchase and storage practices, have resulted in an increased reliance on refrigeration to maximize food safety. Previous research suggests that many domestic refrigerators operate at temperatures exceeding recommendations; however, the results of several studies were determined by means of one temperature data point, which, given temperature fluctuation, may not be a true indicator of actual continual operating temperatures. Data detailing actual operating temperatures and the effects of consumer practices on temperatures are limited. This study has collated the time-temperature profiles of domestic refrigerators in consumer kitchens (n = 43) over 6.5 days with concurrent self-reported refrigerator usage. Overall, the findings established a significant difference (P < 0.05) between one-off temperature (the recording of one temperature data point) and mean operating temperature. No refrigerator operated at ≤5.0°C for the entire duration of the study. Mean temperatures exceeding 5.0°C were recorded in the majority (91%) of refrigerators. No significant associations or differences were determined for temperature profiles and demographics, including household size, or refrigerator characteristics (age, type, loading, and location). A positive correlation (P < 0.05) between room temperature and refrigerator temperature was determined. Reported door opening frequency correlated with temperature fluctuation (P < 0.05). Thermometer usage was determined to be infrequent. Cumulatively, research findings have established that the majority of domestic refrigerators in consumer homes operate at potentially unsafe temperatures and that this is influenced by consumer usage. The findings from this study may be utilized to inform the development of shelf-life testing based on realistic domestic storage conditions. Furthermore, the data can inform the development of future educational interventions to increase safe domestic refrigeration practices.

Increased consumer demand for ready-to-eat (RTE) food products has been documented over recent years, with many consumers seeking convenient food that is appealing, high in quality, perceived to be fresher, and free from preservatives (23, 43). Consequently, preservation processes have become milder, with less use of acids, salt, and chemical additives to control microbiological activity in RTE food products (40). The microbiological safety of such food products is achieved by the combination of innovative technologies, such as modified atmosphere packaging, and refrigeration (6, 37). Given the association of many RTE food products with the psychrotrophic pathogen Listeria monocytogenes, strict temperature control is required to prevent proliferation (18), particularly after food products packed in modified atmosphere packaging are opened.

Many factors have resulted in changes in consumer purchase, storage, and cooking habits. In 2015, 7.7 million people were reported to live in single-person households in the United Kingdom (35), and it is suggested that consumers from single-person households are more likely to cook in bulk. However, men that live alone report that it is not worth cooking a proper meal and that they may utilize RTE foods (30). The purchases of consumers in the United Kingdom are reportedly influenced by in-store promotions that give a discount for purchasing a greater quantity (31), and many consumers are reported to look for resealable packaging in an attempt to reduce food wastage (32). These data may suggest that prolonged storage of food products in the domestic kitchen does occur. This, coupled with consumers wanting to reduce time spent on shopping, food preparation, and cooking, may result in food products being stored in the home for prolonged periods of time (4). Such practices, which result in prolonged storage, may be detrimental for food safety if food products are subject to inadequate storage temperatures.

Stringent measures are in place throughout the food industry supply chain to ensure that food products in the cold chain maintain food safety properties. Food hygiene regulations in the United Kingdom require that food products capable of supporting the growth of pathogens must be kept at temperatures of ≤8.0°C at all stages of the supply chain (15). Although consumers in the United Kingdom rely on food manufacturers and suppliers to ensure that the RTE food products they purchase are safe for consumption, the post-purchase responsibility is in the hands of the consumer to implement safe food storage and handling practices (17). Although domestic refrigeration is not covered by regulations (27), maintenance of the recommended low storage temperatures when food products reach the home is critical (37), and United Kingdom refrigeration recommendations suggest a maximum operating temperature of 5.0°C (16) (the recommendation in the United States is 40°F [4.4°C] (7)). It is reported that refrigeration practices in the domestic kitchen can have a significant impact on the safety of food (25), particularly given the increased reliance on temperature and packaging to control microbiological growth (6). Failing to ensure adequate refrigeration temperatures is reported to be the greatest hazard with RTE foods (6), and temperature control is essential to maintain the safety of minimally processed RTE food products (38).

Several consumer food safety studies have reported that consumer knowledge of safe refrigeration temperatures is lacking (9, 11). Indeed, a lack of consumer awareness and ability to assess safe refrigeration practices may result in unsafe refrigerated-storage temperatures. Previous research has determined that domestic refrigerators often operate at temperatures that are higher than recommended. A recent review established that 21% of consumer food safety studies determined the operating temperatures of consumers' domestic refrigerators (9), and the findings indicated that such temperatures exceeded the national guidelines at rates of 26% in France (27), 47% in the United States (8), 55% in Greece (38), 56% in Wales (41), 59% in Ireland (24), 60% in New Zealand (34), 66% in The Netherlands (39), and 71% in Portugal (5).

Previous research on domestic refrigerator temperatures has been determined using a single temperature reading (10). However, limitations may exist regarding means of single recordings, as refrigeration temperatures are known to fluctuate. Although a one-off temperature (recording of one temperature data point) indicates the temperature at a fixed point in time, it does not capture the temperature range resulting from consumer use and is consequently not a true representation of refrigeration in consumers' domestic kitchens. To date, there is a lack of data detailing the impacts of consumer behavior on refrigeration temperatures and actual temperature profiles, which account for operational temperature fluctuations.

Therefore, this study aims to ascertain the operating temperatures of consumers' domestic refrigerators by means of time-temperature profiling and determination of self-reported practices related to refrigerator usage. Such data will give a better understanding of the risks that may be associated with consumer refrigeration practices in the domestic kitchen. The findings may be utilized to inform the shelf-life testing of RTE food products, allow recommendations to improve domestic refrigerators, and inform the development of food safety risk communication to improve consumer refrigeration practices.

Recruitment. Forty-three consumers from an urban community in South Wales, United Kingdom, were recruited (from a convenience sample using online advertising and posters on community notice boards) and consented to participate in the study. Supermarket vouchers worth £5 were offered as incentives for participation in the research study. Ethical approval was granted from the Cardiff School of Health Sciences (Cardiff Metropolitan University) Research Ethics Committee (reference 2221).

Data collection. To allow for data comparison, one-off operating temperatures at two locations (in a central storage location and in the refrigerator door) for each refrigerator were recorded prior to commencing the time-temperature profiling study, using two calibrated digital handheld battery-powered thermometer probes (P300 handheld thermometer, Industrial Temperature Sensors Co., Kildare, Ireland) with a measuring range of −40 to +200°C and accuracy of ±1.0°C. The probes were left in the refrigerators for ≤5 min or until the temperature readings stabilized.

The time-temperature profiling study was conducted in the consumers' domestic kitchens over a period of six days, during which two data loggers (SL52T self-contained, single-channel, button temperature logger, Signatrol Ltd., Tewkesbury, UK) that record up to 8,000 temperature measurements at a frequency of 1 per min, with a range of −40 to +85°C and accuracy of ±0.5°C, were placed in two locations within the refrigerator (central food storage area and door storage area). In addition, one data logger was placed outside the refrigerator to record the room temperature. Over the 136.5-h time-temperature profiling period, all refrigerator users in each home recorded refrigerator usage in a refrigerator usage log: this included door-opening frequency, the time when the refrigerator was used, and for what reason. Following completion of the temperature profiling, the completed refrigerator usage logs and data loggers were collected from participants' homes, at which point further information regarding the refrigerator was collected, including its age, type, and proximity to a heat source in the kitchen, as well as self-reported refrigeration practices.

Data analysis. Temperature logger data were downloaded using the TempIT-Pro operating software (Signatrol Ltd.). The self-reported data recorded in the refrigerator usage logs were entered into a specifically designed database (Microsoft Office Access 2010, Microsoft, Redmond, WA) to store and select data for analysis. Profiled time-temperature data were analyzed using Microsoft Excel 2010. Statistical analysis was conducted using SPSS Statistics 20 (IBM Software Group, Chicago, IL).

Domestic refrigerator operating-temperature profiles. The temperature profiling was completed in 43 domestic kitchens, and 24,576 data points were recorded in each domestic kitchen. The temperature was recorded each minute for 136.5 h from two locations within the refrigerator (central food storage area and door storage area), and the room temperature of the kitchen was also recorded. The domestic refrigerator operating temperatures recorded during time-temperature profiling ranged from −1.72 up to 17.9°C.

The profiling findings indicate that 91% of refrigerators had mean operating temperatures that exceeded recommendations (>5.0°C) (Fig. 1); only four refrigerators (9%) were identified as having mean temperatures that were 5.0°C or below in both central and door locations of the refrigerator. The mean operating temperatures ranged from 1.81 to 11.38°C (mean, 5.8°C; standard deviation [SD], 1.9). The majority (98%) of refrigerators had lower mean temperatures in the center of the refrigerator than in the door, with differences in temperature ranging between 0.1 and 4.5°C (Table 1).

FIGURE 1.

Frequency curve of mean operating temperatures of domestic refrigerators (n = 43) from the data logger study in consumer domestic kitchens.

FIGURE 1.

Frequency curve of mean operating temperatures of domestic refrigerators (n = 43) from the data logger study in consumer domestic kitchens.

Close modal
TABLE 1.

Temperature ranges in domestic refrigerators

Temperature ranges in domestic refrigerators
Temperature ranges in domestic refrigerators

Although 9% of refrigerators had mean temperatures that were within United Kingdom recommended parameters in both central and door storage areas, no refrigerators (door and central location) were discovered to operate at 5.0°C or below for the duration of the profiling study (see Table 2). The majority of refrigerator door storage areas (70%) and central locations (40%) were operating at unsafe temperatures for the duration of the profiling study (100% of the time). Cumulatively, 40% of refrigerators had both door and central operating temperatures of >5.0°C for the duration of study. It was calculated that 5% of refrigerators operated within United Kingdom recommended temperatures for 80% of the profiling duration.

TABLE 2.

Percentages of time domestic kitchen refrigerators were operating at temperatures exceeding recommendationsa

Percentages of time domestic kitchen refrigerators were operating at temperatures exceeding recommendationsa
Percentages of time domestic kitchen refrigerators were operating at temperatures exceeding recommendationsa

Forty-one percent of refrigerators were in one-person households. Statistical analysis determined that no significant differences (P > 0.05) existed between refrigeration temperatures or mean daily temperature changes and participant demographics, including household type (one-person or multiple-person household). Similarly, no correlation existed (P > 0.05) between daily temperature changes or mean operating temperature and the number of household occupants. A chi-square test for independence indicated no significant association between household size (one-person household or multiple-person household) and potentially safe or unsafe mean operating temperatures (≤5.0 or >5.0°C).

The reported ages of domestic refrigerators in consumer kitchens ranged from 4 months to 30 years. No relationships were determined between the reported ages of the 43 domestic refrigerators and performance, including temperature changes and times taken to return to previous operating temperature after putting food shopping in to the refrigerator (P > 0.05) and mean operating temperature of refrigerators (P > 0.05). A chi-square test for independence indicated no significant association between the age group of refrigerators (<5, 5 to 10, or >10 years) and potentially safe or unsafe mean operating temperatures (≤5.0 or >5.0°C).

The majority (70%) of refrigerators were freestanding and nonintegrated, while the remainder were built-in integrated appliances. No significant difference (P > 0.05) in the mean operating temperatures existed between refrigerator types, and no associations were determined between refrigerator types and potentially safe or unsafe mean operating temperatures (≤5.0 or >5.0°C). The majority (70%) of refrigerators were combined refrigerator-freezers, of which the majority (63%) had the freezer located below the refrigerator, 13% had freezers above the refrigerator, and 23% had freezer compartments contained within the refrigerator. No significant differences (P > 0.05) were determined for the mean operating temperature or daily temperature changes according to the location of the freezer, and no associations (P > 0.05) were determined between freezer location and potentially safe or unsafe mean operating temperatures (≤5.0 or >5.0°C).

Twelve percent of refrigerators were observed to be nearly empty at the start and completion of the study, 12% were at full capacity, and 76% were half-loaded with food products. No significant differences were determined (P > 0.05) between the loading of refrigerators and the operating temperatures, and no associations were determined between loading and safe or unsafe temperatures.

A Pearson product-moment correlation coefficient determined a moderate positive correlation between the one-off recorded temperatures and the mean operating temperatures of refrigerators (r = 0.35, n = 43, P < 0.05), with increased one-off temperatures associated with increased mean time-temperature profiling temperatures. However, a paired-sample t test determined that there was a statistically significant difference between the one-off recorded temperatures (mean, 4.94; SD, 2.30) and the mean operating temperature of refrigerators during the time-temperature profiling [mean, 5.83; SD, 1.94; t (42) = −2.39; P < 0.05 (two-tailed)]. The mean difference was −0.89°C, with a 95% confidence interval the eta-squared statistic (0.01) indicated a small effect.

As seen here, refrigerators do not operate at a fixed temperature; fluctuations do occur, and compared to commercial refrigerators, domestic refrigerators are known to have smaller compressors, single-layer insulation, and lower airflow, which may result in a longer temperature recovery time (29, 42). A trend in the changes of room and refrigerator temperatures during the course of 24 h was noted. Temperatures were found to be at their lowest between the hours of 0500 and 0700 and to increase gradually during the course of the day, and the highest temperatures were recorded between the hours of 1700 and 2100, peaking at around 1900. Additionally, trends were determined between refrigerator operating temperatures and room temperatures; analysis using a scatterplot determined a positive relationship between the room temperatures and the operating temperatures of refrigerator central storage areas (Fig. 2). A Pearson product-moment correlation coefficient determined a positive correlation between mean room temperatures and mean refrigerator door operating temperatures (r =0.415, n =43, P < 0.005); similarly, mean room temperatures and central operating temperatures of domestic refrigerators had a positive correlation (r = 0.303, n = 43, P < 0.05). The findings indicate that the greater the room temperature, the greater the refrigerator operating temperature. Although nine of the refrigerators (21%) were observed or were self-reported to be in direct sunlight, no significant differences in mean temperatures or daily temperature changes according to sunlight positioning were determined (P > 0.05).

FIGURE 2.

Scatter graph indicating the relationship between mean room temperatures and refrigerator central storage area mean operating temperatures (°C) in consumer domestic kitchens (n = 43).

FIGURE 2.

Scatter graph indicating the relationship between mean room temperatures and refrigerator central storage area mean operating temperatures (°C) in consumer domestic kitchens (n = 43).

Close modal

Self-reported refrigeration practices. The temperature logger data and the self-reported refrigerator usage logs kept by the participants during the course of time-temperature profiling study were analyzed to determine any trends in temperature fluctuation according to reported usage. A linear relationship was determined between daily door opening frequencies and temperature changes, as indicated in Figure 3. A small positive correlation was determined between daily temperature fluctuations (difference between lowest and highest recorded temperature) and self-reported door opening frequencies (r =0.29, n =258, P < 0.0005).

FIGURE 3.

Scatter graph indicating the relationship between daily temperature fluctuation (°C) (n = 258) and self-reported door-opening frequency in consumer domestic kitchens (n = 43) over a 6-day period.

FIGURE 3.

Scatter graph indicating the relationship between daily temperature fluctuation (°C) (n = 258) and self-reported door-opening frequency in consumer domestic kitchens (n = 43) over a 6-day period.

Close modal

Many consumers indicated confusion between actual operating temperature and the setting on the refrigerator dial. No statistically significant associations (P > 0.05) were determined between mean operating temperatures and refrigerator dial settings. Reported ownership and usage of temperature thermometers was determined to be lacking, with only 23% of consumers reporting that they owned a thermometer and 13% of refrigerators having built-in temperature thermometers or gauges. During the time-temperature profiling study, 13% of participants reported having checked the operating temperature of the home refrigerator; however, no participants reported changing the refrigerator operating temperature or adjusting the dial during the period of the study. No significant differences (P > 0.05) were determined in potentially safe or unsafe mean operating temperatures (≤5.0 or >5.0°C) according to reported ownership of thermometers, having built-in temperature indicators, or reporting having checked the temperature. A matter of concern is that the mean operating temperatures in refrigerator door storage areas were determined to be significantly greater (U = 96.5, z = −1.998, P < 0.05, r =−0.31) in homes that were observed or self-reported to have thermometers (mean, 8.9°C; SD, 1.5) than in those without thermometers (mean, 7.3°C; SD, 2.2).

Changes in the operating temperatures of refrigerators were determined at the times consumers reported putting food in the refrigerator after returning home from food shopping. On average, a temperature change of 1.92°C (SD, 1.6) at the center of the refrigerator was noted; however, this ranged from an increase of 7.05°C to a decrease of 2.01°C. A general trend was observed in that the time to reach the maximum temperature change after putting food shopping into the refrigerator was much shorter than the time taken to return to the original temperature. On average, the maximum temperature was reached within 18 min of the self-reported time when food was placed in the refrigerator after shopping; this ranged from 1 min up to 65 min. The times to return to the original temperature were between 2 and 782 min (13 h), with an average of 143 min (2.4 h). It was determined that 17% of refrigerators took >5 h to recover to the original temperature after food from shopping was placed into the refrigerator, and three of these refrigerators took over 10 h. Although no significant differences (P > 0.05) were determined in temperature recovery according to refrigerator age, refrigerators with a temperature recovery of >5 h were on average older (average age of 11 years) than refrigerators with a temperature recovery of <5 h (average age of 6.6 years). Furthermore, the average temperature recovery time of refrigerators aged >5 years was longer (2.9 h) than that of refrigerators aged <5 years (1.6 h).

The method of putting food in the refrigerator after shopping varied among participants and were categorized into three methods, as follows: (i) continuously reopening the refrigerator door and placing a single or a few items into refrigeration each time, (ii) opening the refrigerator door when starting to put food away and leaving it open until all food products had been placed in the refrigerator, and (iii) organizing all items for refrigeration prior to opening the refrigerator and placing items in all at once. The majority of participants reported usually (and had done so during the time-temperature profiling) organizing all of the food products that needed to be refrigerated together in one area of the kitchen while going through purchased products and placing all of them into refrigeration at the same time. The frequencies of the self-reported methods of putting food into the refrigerator after shopping are detailed in Table 3. No significant differences were determined (P > 0.05) between the mean temperature changes and the times taken to return to original operating temperature after putting food in the refrigerator according to the reported method of putting food away.

TABLE 3.

Comparison of temperature changes according to self-reported methods of putting food into the refrigerator after shoppping

Comparison of temperature changes according to self-reported methods of putting food into the refrigerator after shoppping
Comparison of temperature changes according to self-reported methods of putting food into the refrigerator after shoppping

Best-case and worse-case examples of post-shopping refrigerator temperature fluctuations have been selected to illustrate the impact of self-reported refrigerator usage on temperature. The time-temperature profile of the refrigerator in the domestic kitchen of participant MP014 and self-reported practices when putting food into the refrigerator after shopping are indicated in Figure 4, where it can be seen that the temperature increased by 2.51°C 6 min after the participant reported putting food in to the refrigerator; the refrigerator took 36 min to return to the temperature it was operating at prior to putting the food into the refrigerator. As indicated in Figure 5, the time taken to return to previous operating temperature after shopping was longer for participant MP056, where a greater number of refrigerator door openings were reported and the refrigerator operating temperature increased by 7.00°C after putting food into the refrigerator; a further 17 reported door openings occurred, and the refrigerator took 782 min to return to previous operating temperature. The findings illustrate both that consumer practices influence operating temperature and that there is a need for improved temperature recovery in domestic refrigerators.

FIGURE 4.

Time-temperature profile of self-reported refrigerator usage when putting food into the refrigerator after shopping (participant MP014).

FIGURE 4.

Time-temperature profile of self-reported refrigerator usage when putting food into the refrigerator after shopping (participant MP014).

Close modal
FIGURE 5.

Time-temperature profile of self-reported placing of food into the refrigerator after shopping (participant MP056).

FIGURE 5.

Time-temperature profile of self-reported placing of food into the refrigerator after shopping (participant MP056).

Close modal

The data determined through this time-temperature profiling study are crucially important, as they not only give an understanding of actual temperature fluctuations in domestic refrigerators but give in-depth, more accurate insight into refrigeration operating conditions and information on the impact of consumer behavior on refrigeration temperature and subsequent microbiological food safety.

Domestic refrigerator operating-temperature profiles. The domestic refrigerator operating temperatures profiled in this study ranged between −1.7 and 17.9°C; the maximum temperature determined is greater than any determined in previous research (2, 24, 27). Previous analysis of the temperature performance of domestic refrigerators in consumer domestic kitchens has been conducted in France, and the findings indicated that no relationship existed between thermometer and data logger temperatures (27). Although correlations were identified in this study between thermometer and mean data logger temperatures, statistical analysis determined that the one-off thermometer temperatures were significantly different from the mean operating temperatures from the time-temperature profiling. This suggests that one-off temperatures are not indicative of actual operating temperatures and highlights the need for such time-temperature profiling data over prolonged periods of time.

Previous research has indicated that ≤10% of refrigerators in domestic kitchens operate at between 0.0 and 5.0°C for 90% of the time (13, 22). However, in this study, no refrigerators operated at the recommended temperatures for 90% of the time-temperature profiling study. The results from this research showed that only 5% of refrigerators operated at the recommended temperatures for 80% of the profiling duration. Furthermore, the findings from this study indicate that 40% of refrigerators had both central and door temperatures that were higher than recommended (5.0°C) for the entire duration of the study, indicating substantial potential for microbiological growth in foods stored in such refrigerators.

As anticipated, the central storage areas of the refrigerators had lower mean temperatures than the refrigerator doors. Previous research suggests that temperatures vary at different locations within the refrigerator, with the top of the refrigerator reported to be the warmest location in the majority of refrigerators (22).

Previous older adult consumer food safety research has suggested links between the reported ages of refrigerators and their operating temperatures (10), findings which may have been influenced by the consumer group selected for the study, the larger proportion of refrigerators aged 10 years or older, and the use of one-off operating temperatures. However, in the present study, no relationships were observed between the reported ages of the 43 domestic refrigerators and their refrigeration performance as determined by the profiled temperature data. Previous research has not determined that refrigerator characteristics such as age influence refrigerator temperature (22) or temperature fluctuation (2).

The room temperatures external to refrigerators and the positioning of refrigerators may influence the internal operating temperatures. The findings from this study indicated a positive correlation between room temperatures and operating temperatures, whereby increased room temperature was associated with increased refrigerator operating temperature. This finding is in concurrence with Rahman et al. (36), who also report that the internal temperature of domestic refrigerators is influenced by the kitchen temperature. However, other research has indicated the contrary, whereby factors such as ambient temperature have no effect on the refrigeration temperature (22). Given the findings from this study indicating an impact of kitchen temperature upon refrigeration temperature, there is a need to place refrigerators in cooler areas to maximize the chance of maintaining a temperature of ≤5.0°C; these findings also indicate the need to improve the insulation of domestic refrigerators.

Factors such as household size and refrigerator type may affect temperature performance and have implications for food safety in the domestic setting. However, previous studies have reported no significant relationships between temperatures and household demographic characteristics, including the number of persons in the household (28). Similarly, in this study, it was determined that no significant differences or associations existed between temperature performance and household size. James and Evans (22) determined that refrigerator temperatures varied according to refrigerator types; however, in this study, refrigerator types (freestanding nonintegrated or built-in integrated and location of freezer) and temperatures had no significant associations.

Self-reported refrigeration practices. James and Evans (22) stated a need to determine the influence of loading patterns and door opening frequencies on refrigerator temperature performance, and previous research suggests that refrigerator loading influences the temperature performance of domestic refrigerators (20, 21). Temperature profiles from consumer refrigerators were analyzed in this study in accordance with recorded usage, including loading and opening frequency. No significant associations were determined between refrigerator loading and temperature performance; however, a positive correlation was determined between daily temperature fluctuation and self-reported door-opening frequency. Indeed, this corroborates clustering analysis data that suggest that refrigerators with greater door-opening frequencies have higher temperatures than those with less frequent door openings (36); such factors may have implications for food safety, as increased temperatures may cause microbial proliferation.

Previous studies have determined that self-reported ownership of a refrigerator thermometer was low in a number of countries, including European countries and the United States. No consumers in Malta (19), 19% of United Kingdom consumers (33), 23 to 24% of Irish consumers (14, 24), and 11 to 24% of U.S. consumers (1, 26) reported owning or were found to have a thermometer in their refrigerators. In concurrence with these findings, less than a quarter of consumers in this study were observed to have or reported owning a refrigerator thermometer. Reported ownership of a refrigerator thermometer does not necessarily translate into safe refrigeration practices, as less than half (46%) of consumers in an American study reported awareness of how to monitor for the correct temperature (3). However, ownership of a temperature thermometer does indicate motivation and intention to monitor and maintain safe refrigerator temperature. It may not be consumers' intention to subject food to unsafe storage conditions, as improper refrigeration can be the accidental result of mechanical faults, the result of consumer ignorance, or an intentional practice to save on energy costs. Although no associations existed between thermometer ownership and potentially safe or unsafe mean operating temperatures, the mean operating temperatures in refrigerator door storage areas were determined to be significantly higher in homes that were observed to have or reported having thermometers (mean, 8.9°C) than in those without thermometers (mean, 7.3°C). This may suggest that consumers are aware of the importance of refrigeration to ensure food safety but are unaware of the recommended temperatures or fail to adequately use thermometers in refrigerators to assess safe refrigeration temperatures.

As established in a previous study by Evans and Redmond (11), many consumers have expressed confusion or misunderstanding between the actual operating temperature and the setting on the refrigerator dial. In this study, refrigerator dial settings were not associated with mean operating temperatures. This agrees with previous research that determined that no direct relationship exists between refrigerator settings and refrigeration temperatures (36). Flynn et al. (12) observe that thermostat dials are often not adjusted by consumers to achieve the minimum temperature, and Laguerre et al. (27) indicate that no relationships exist between refrigeration temperatures and refrigerator settings. Consumer reliance on the setting on the refrigerator dial as an indicator of safe operating temperature may result in unintentional unsafe food storage conditions.

Previous research suggests that increased mean temperatures in domestic refrigerators correspond to shopping days (36), but in this study, no statistically significant differences were determined between the temperature change after putting away food from shopping and the time taken to return to the temperature before putting food away or between the reported methods of putting food in the refrigerator after returning home from food shopping. Despite this, the number of door openings after food shopping was seen to affect the time taken to return to the operating temperature of the refrigerator prior to putting food in, which indicates that such practices may have implications for food safety.

In conclusion, prior to conducting this time-temperature profiling study, data detailing the effects of consumers' self-reported practices on domestic refrigerator operating temperature profiles were limited. Cumulatively, the majority of domestic refrigerators in this study operated at temperatures exceeding those recommended for safe refrigerated storage, which might affect the microbial safety of food products.

This study has provided novel data comparing refrigerator temperature profiles and consumers' self-reported refrigeration practices and further elucidated the impact of consumer behavior upon refrigerator operating temperatures. Indeed, temperature fluctuations determined to result from increased door opening when loading with food after shopping resulted in an increased temperature and a prolonged time to return to the previous temperature, thus increasing the risk of microbiological growth in food products; such growth in RTE food products may have consequences for domestic food safety.

The cumulative findings of this study provide data that can inform the development of shelf-life testing based upon realistic refrigerated storage conditions in the domestic kitchen. Furthermore, the data can also be utilized to inform the development of risk communication strategies to raise awareness of correct refrigeration temperatures and educate and potentially motivate consumers to implement safe refrigeration practices. Given the determination of fluctuating temperatures, there is a need for laboratory trials to replicate the fluctuating domestic refrigeration at temperatures exceeding recommendations to establish the impact on psychrotrophic pathogens like L. monocytogenes in RTE food products.

It is recommended that refrigerator design be improved, including strengthened seals, better insulation, self-closing doors, and temperature recovery systems, which may help consumers to implement safe refrigeration practices in the home and reduce the risks associated with foodborne illnesses like listeriosis.

This study was supported by research funds from the Vice Chancellor's Doctoral Award from Cardiff Metropolitan University. The authors wish to acknowledge Professor Louise Fielding (1968 to 2013) for her support throughout the project.

1.
Albrecht
,
J.
1995
.
Food safety knowledge and practices of consumers in the U.S.A
.
J. Consumer Stud. Home Econ
.
19
:
119
134
.
2.
Albrecht
,
J. A.
, and
C.
Larvick
.
2007
.
Refrigerator practices of participants in the Meals on Wheels Program
.
Food Prot. Trends
27
:
672
677
.
3.
American Dietetic Association and ConAgra Foods
.
1999
.
Home food safety benchmark survey
.
4.
Audit Commission for Local Authorities and the National Health Service in England and Wales
.
1990
.
Safer food: local authorities and the Food Safety Act, 1990. Occasional papers 15
.
Audit Commission for Local Authorities and the National Health Service in England and Wales
,
London
.
5.
Azevedo
,
I.
,
M.
Regalo
,
C.
Mena
,
G.
Almeida
,
L.
Carneiro
,
P.
Teixeira
,
T.
Hogg
, and
P.
Gibbs
.
2005
.
Incidence of Listeria spp. in domestic refrigerators in Portugal
.
Food Control
16
:
121
124
.
6.
Brackett
,
R.
1992
.
Microbiological safety of chilled foods: current issues
.
Trends Food Sci. Technol
.
3
:
81
85
.
7.
Centers for Disease Control and Prevention
.
2011
.
Listeria (listeriosis). Prevention
.
Available at: http://www.cdc.gov/listeria/prevention.html. Accessed 8 November 2013
.
Centers for Disease Control and Prevention
,
Atlanta
.
8.
Dharod
,
J.
,
R.
Perez-Escamilla
,
S.
Paciello
,
K.
Venkitanarayanan
,
A.
Bermudez-Millan
, and
G.
Damio
.
2007
.
Critical control points for home prepared ‘chicken and salad’ in Puerto Rican households
.
Food Prot. Trends
27
:
544
575
.
9.
Evans
,
E.
, and
E.
Redmond
.
2014
.
Behavioural risk factors associated with listeriosis in the home: a review of consumer food safety studies
.
J. Food Prot
.
77
:
510
521
.
10.
Evans
,
E.
, and
E.
Redmond
.
2015
.
Analysis of older adults' domestic kitchen storage practices in the United Kingdom: identification of risk factors associated with listeriosis
.
J. Food Prot
.
78
:
738
745
.
11.
Evans
,
E.
, and
E.
Redmond
.
2016
.
Older adult consumer knowledge, attitudes and self-reported storage practices of ready-to-eat food products and risks associated with listeriosis
.
J. Food Prot
.
79
:
263
272
.
12.
Flynn
,
O.
,
I.
Blair
, and
D.
McDowell
.
1992
.
The efficiency and consumer operation of domestic refrigerators
.
Int. J. Refrig
.
15
:
307
312
.
13.
Food and Drink Federation and Institution of Environmental Health Officers
.
1994
.
FOODLINK: national food survey
.
Food and Drink Federation and the Institution of Environmental Health Officers
,
London
.
14.
Food Safety Authority of Ireland
.
1998
.
Public knowledge and attitudes to food safety in Ireland
.
Food Safety Authority of Ireland
,
Dublin
.
15.
Food Standards Agency
.
2016
.
Guidance on Temperature Control Legislation in the United Kingdom
.
Food Standards Agency
,
London
. .
16.
Food Standards Agency and Department of Health
.
2008
.
Listeria—keeping food safe
.
Food Standards Agency and Department of Health
,
London
. .
17.
Foodborne Disease Working Party for the Communicable Diseases Network Australia and New Zealand
.
1997
.
Foodborne disease: towards reducing foodborne illness in Australia. Technical report series no. 2
.
Communicable Diseases Network Australia and New Zealand
,
Canberra, Commonwealth of Australia
.
18.
Garrido
,
V.
2009
.
The problem of listeriosis and ready-to-eat products: prevalence and persistence
.
In
Current Research Topics in Applied Microbiology and Microbial Biotechnology
,
Proceedings of the II International Conference on Environmental, Industrial and Applied Microbiology
,
28 November to 1 December 2007
,
University of Seville, Spain
.
World Scientific Publishing Company
,
Singapore
.
19.
Gauci
,
C.
2000
.
Consumer awareness to Salmonellosis
.
M.S. thesis, University of Malta
.
20.
James
,
S.
, and
J.
Evans
.
1992
.
The temperature performance of domestic refrigerators
.
Int. J. Refrig
.
15
:
313
319
.
21.
James
,
S.
,
J.
Evans
, and
C.
James
.
2008
.
A review of the performance of domestic refrigerators
.
J. Food Eng
.
87
:
2
10
.
22.
James
,
S. J.
, and
J.
Evans
.
1992
.
Consumer handling of chilled foods: temperature performance
.
Int. J. Refrig
.
15
:
299
306
.
23.
Jekanowski
,
M.
,
J.
Binkley
, and
J.
Eales
.
2001
.
Convenience, accessibility, and the demand for fast food
.
J. Agric. Resour. Econ
.
26
:
58
74
.
24.
Kennedy
,
J.
,
V.
Jackson
,
I.
Blair
,
D.
McDowell
,
C.
Cowan
, and
D.
Bolton
.
2005
.
Food safety knowledge of consumers and the microbiological and temperature status of their refrigerators
.
J. Food Prot
.
68
:
1421
1430
.
25.
Kilonzo-Nthenge
,
A.
,
F.-C.
Chen
, and
S.
Godwin
.
2008
.
Occurrence of Listeria and Enterobacteriaceae in domestic refrigerators
.
J. Food Prot
.
71
:
608
612
.
26.
Kosa
,
K.
,
S.
Cates
,
S.
Karns
,
S.
Godwin
, and
D.
Chambers
.
2007
.
Consumer knowledge and use of open dates: results of a web-based survey
.
J. Food Prot
.
70
:
1213
1219
.
27.
Laguerre
,
O.
,
E.
Derens
, and
B.
Palagos
.
2002
.
Study of domestic refrigerator temperature and analysis of factors affecting temperature: a French survey
.
Int. J. Refrig
.
25
:
653
659
.
28.
Marklinder
,
I.
,
M.
Lindblad
,
L.
Eriksson
,
A.
Finnson
, and
R.
Lindqvist
.
2004
.
Home storage temperatures and consumer handling of refrigerated foods in Sweden
.
J. Food Prot
.
67
:
2570
2577
.
29.
Marvell Scientific
.
2015
.
Refrigeration for health, science and industry. The difference is in the U.L. listing. Benefits and features of commercial U.L. vs. household U.L.
Marvel Scientific
,
Richmond, IN
. .
30.
Mintel Group Ltd
.
2009
.
Meals-for-one—UK
.
Mintel Group Ltd.
,
London
.
Available at: http://store.mintel.com/meals-for-one-uk-december-2009. Accessed 25 November 2015
.
31.
Mintel Group Ltd
.
2012
.
Cheese—UK—October 2012
.
Mintel Group Ltd.
,
London
.
Available at: http://store.mintel.com/cheese-uk-october-2012. Accessed 8 November 2014
.
32.
Mintel Group Ltd
.
2016
.
Food Packaging Trends—UK—January 2016
.
Mintel Group Ltd.
,
London
. .
33.
National Consumer Council and Great Britain Food Safety Directorate
.
1991
.
Time-temperature indicators: research into consumer attitudes and behaviour
.
Food Safety Directorate, Ministry of Agriculture, Fisheries and Food
,
London
.
34.
O'Brien
,
G. D.
1997
.
Domestic refrigerator air temperatures and the public's awareness of refrigerator use
.
Int. J. Environ. Health Res
.
7
:
141
148
.
35.
Office for National Statistics
.
2015
.
Statistical bulletin. Families and households: 2015. Office for National Statistics, London
. .
36.
Rahman
,
S.
,
N.
Mohd Sidik
,
M.
Hassan
,
T.
Mohd Rom
, and
I.
Jauhari
.
2005
.
Temperature performance and usage conditions of domestic refrigerator-freezers in Malaysia
.
Trans. Hong Kong Inst. Eng
.
12
:
30
35
.
37.
Richmond
,
M.
1990
.
The microbiological safety of food. Part 1. Report of the Committee on the Microbiological Safety of Food to the Secretary of State for Health, the Minister of Agriculture, Fisheries and Food and the Secretaries of State for Wales, Scotland and Northern Ireland
.
Her Majesty's Stationery Office
,
London
.
38.
Sergelidis
,
D.
,
A.
Abrahim
,
A.
Sarimvei
,
C.
Panoulis
,
P.
Karaioannoglou
, and
C.
Genigeorgis
.
1997
.
Temperature distribution and prevalence of Listeria spp. in domestic, retail and industrial refrigerators in Greece
.
Int. J. Food Microbiol
.
34
:
171
177
.
39.
Terpstra
,
M.
,
L.
Steenbekkers
,
N.
de Maertelaere
, and
S.
Nijhuis
.
2005
.
Food storage and disposal: consumer practices and knowledge
.
Br. Food J
.
107
:
526
533
.
40.
Theron
,
M.
, and
J.
Lues
.
2007
.
Organic acids and meat preservation: a review
.
Food Rev. Int
.
23
:
141
158
.
41.
Worsfold
,
D.
, and
C.
Griffith
.
1997
.
Assessment of the standard of consumer food safety behavior
.
J. Food Prot
.
60
:
399
406
.
42.
WRAP
.
2010
.
Part 1: Insights around the domestic refrigerator. Reducing food waste through the chill chain
.
WRAP
,
Banbury, United Kingdom
. .
43.
Yu
,
L.
, and
G.
Hailu
.
2010
.
Household demand for convenience chicken meat products in Canada
.
Consumer and Market Demand Agricultural Policy Research Network. Department of Food, Agricultural and Resource Economics (FARE), University of Guelph
. .