Effect of hot and cold fluids on oral temperatures

Nursing textbooks have differed in recommendations for length of time (10 to 30 minutes) to wait to take the patient's temperature after he has ingest hot or cold food or fluids. It would appear either that inaccurate temperatures are recorded or that an excessive amount of time has been recommended. The purpose of this study was to determine the length of time required for the oral temperature to return to an established baseline after ingestion of hot and cold fluids. The sample consisted of 13 female and 12 male subjects with an age range of 20 to 52 years. All subjects were afebrile. Hot chocolate at a mean temperature of 131 degrees Fahrenheit and iced water at 34 degrees Fahrenheit were ingested by the subjects after a baseline temperature was established. Oral temperatures were recorded every 15 seconds by means of a pretested telethermometer until a constant temperature was maintained for a 10 minute period. In analyzing results, a variation of 0.2 degree was not considered significant because of error of 0.2 degree in reading a clinical thermometer could be made and because medical diagnosis and therapy do not depend on this minute variation. Means and standard deviations of the time required for attainment of point 0.2 degree from the terminal baseline were determined. The assumption the 98 percent (two standard deviations) of people's temperature would reach that point in 26.9 minutes after ingestion of hot fluids and 27.80 minutes after cold fluids was made. The implication for nursing practice to drawn from this study is that to ensure accuracy of temperature measurement, the nurse should wait 30 minutes before recording the oral temperature.

EFFECT OF HOT AND COLD FLUIDS ON ORAL TEMPERATURES by by Mary Almina Thomas A thesis the faculty of the the University A thesis submitted to the of Utah in partial fulfillment of the the requirements for the the degree of for Master of Science Department of Nursing University of Utah June 1969 1969 This Thesis for the Master of Science Degree by Mary Almina Thomas has been approved May Chairman, Supervisory sory Commlttee Supervisory Committee Supe 1969 Committee Y2Y=/Y?#� � ACKNOWLEDGMENTS Sincere thanks are expressed to the writer's supervisory committee and to Dr. Maxine Cope for their help and encouragement, to Dr. Elaine Dyer for the use of her equipment and to the patients and nursing personnel on 2 West, Latter-day Saints Hospital, for their cooperation and interest. iii TABLE OF CONTENTS Page LIST OF TABLES . . v LIST OF FIGURES • • vi vii ABSTRACT CHAPTER I. II. METHOD.. III . RESULTS. IV. . . . . . . INTRODUCTION. •.. • • . . . . •. • • • . . . DISCUSSION . . • • • . . . • . REFERENCES .. APPENDIX A . • . . • • 1 4 10 20 22 .••.. • 23 REGRESSION ANALYSIS LINE FOR ORAL PROBE. 24 REGRESSION ANALYSIS LINE FOR RECTAL PROBE. 25 VITA . • • . 26 iv LIST OF TABLES Page TABLE I. RESPONSES OF SUBJECTS DRINKING HOT FLUIDS 11 II. MEANS OF AMOUNT OF CHANGE AND RETURN TO BASELINE FOR SUBJECTS DRINKING HOT FLUIDS IN SEX AND AGE CATEGORIES . . • • . . . • 12 RESPONSES OF SUBJECTS DRINKING COLD FLUIDS. 15 MEANS OF AMOUNT OF CHANGE AND RETURN TO BASELINE FOR SUBJECTS DRINKING COLD FLUIDS IN SEX AND AGE CATEGORIES • • • . 16 RETURN TIMES FOR SUBJECTS DRINKING BOTH HOT AND COLD FLUIDS • . • • • • • • . • 19 III. IV. V. v LIST OF FIGURES Page FIGURE 1. 2. RECOVERY PROCESS IN SUBJECTS DRINKING HOT FLUIDS 14 RECOVERY PROCESS IN SUBJECTS DRINKING COLD FLUIDS. 18 .................... . . .. . . . . .. . . .. .. . . . vi ABSTRACT Nursing textbooks have differed in recommendations for length of time (10 to 30 minutes) to wait to take the patient's temperature after he has ingested hot or cold food or fluids. It would appear either that inaccurate temperatures are recorded or that an excessive amount of time has been recommended. The purpose of this study was to deter- mine the length of time required for the oral temperature to return to an established baseline after ingestion of hot and cold fluids. The sample consisted of 13 female and 12 male subjects with an age range of 20 to 52 years. All subjects were afebrile. Hot chocolate at a mean temperature of 131 degrees Fahrenheit and iced water at 34 degrees Fahrenheit were ingested by the subjects after a baseline temperature was established. Oral temperatures were recorded every 15 seconds by means of a pretested telethermometer until a constant temperature was maintained for a 10 minute period. In analyzing results, a variation of 0.2 degree was not considered significant because an error of 0.2 degree in vii reading a clinical thermometer could be made and because medical diagnosis and therapy do not depend on this minute variation. Means and standard deviations of the time re- quired for attainment of a point baseline were determined. 0~2 degree from the terminal The assumption that 98 percent (two standard deviations) of people's temperatures would reach that point in 26.9 minutes after ingestion of hot fluids and in 27.8 minutes after cold fluids was made. The implication for nursing practice to be drawn from this study is that to ensure accuracy of temperature measurement, the nurse should wait 30 minutes before recording the oral temperature. viii CHAPTER I INTRODUCTION The importance of body temperature measurement is indicated by its inclusion by the medical profession as one of the vital signs used as an indicator of health status. Fever is considered to be a sign of the presence and/or severity of a disease process. Accuracy of temperature measurement is, thus, of importance in nursing practice. DuBois (1948) reported a study of the differences in temperature in different parts of the body. The variations were greatest in the areas most exposed to the environment and least in the interior regions of the body. The mouth reflects the interior body temperature and is considered to be the most practical and most commonly used site. However, the mouth is easily affected by local changes in heat loss or gain, such as by hot and cold foods and/or fluids, air temperature, breathing and talking (Selle, 1952). Nursing textbooks have differed rather widely in recommendations regarding the amount of time (10 to 30 minutes) to wait to take the oral temperature after the patient had been smoking, ingested hot or cold food or fluids, or had been chewing gum (Montag and Filson, 1953; Price, 1965; Nordmark and Rohweder, 1967; Rasmussen, 1967). These guidelines have 2 left the nurse practicioner a wide range (20 minutes) from which to choose to wait before measuring the patient's oral temperature. It would appear that either inaccurate tem- peratures are recorded or that an excessive amount of time has been recommended. Brim and Chandler (1948) studied the effects of hot tea and coffee, iced water and lemonade, cigarettes and chewing gum on the oral temperatures of 200 subjects with 50 subjects in each of the four groups. Subjects were allowed to drink fluids, smoke cigarettes or chew gum in their customary manner. Temperatures were taken before, immediately after and at five minute intervals after completion of the activity until the control temperature was regained. Clinical ther- mometers were used to record temperatures for a three minute period, after which they were removed, shaken down and reinserted after two minutes. They found, in subjects drinking hot fluids, from 20 to 80 minutes were required for return to the original temperature. After drinking cold fluids, the majority of the subjects' temperatures returned to the baseline level within five to 10 minutes, but 32 of these temperatures had gone beyond the baseline by the end of 20 minutes. Chewing gum and cigarette smoking both raised and lowered the temperature; however, chewing gum appeared to have a more prolonged effect on the temperature. A few other nursing studies concerning the hot and cold 3 influences on oral temperatures have confirmed changes (Verhonick and Werley, 1963; Woodman, Parry and Simms, 1967), but they do not attempt to estimate the period of variation from the baseline temperature. The present investigator theorized that in the Brim and Chandler study that the recovery process, in the case of hot fluids, might not have been recorded due to the construction of clinical thermometers which prevents the mercury from receding into the bulb until forceably shaken down. No mention of this possibility is made by Brim and Chandler. A further limitation of their study seemed to be the apparent failure to control the temperature and rate of ingestion of the fluids. Neither was mention made of the possible influences of the environmental temperature, the sex and ages of the subjects. The present study was undertaken to replicate the hot and cold fluid phase of Brim and Chandler's study and to include the environmental, sex, and age components. By deter- mining the exact length of time for the return of the temperature to the baseline, it was hoped to be able to offer the nurse practicioner more definite guidelines for accurate oral temperature measurement than are currently available. CHAPTER I I METHOD The study was conducted in a 450 bed private general hospital located in Salt Lake City, utah. The subjects were selected from in-patients who were not in a febrile state or did not have a pathological condition in which fever was expected. The total sample consisted of 12 male and 13 female subjects ranging in age from 20 to 52 years. Six subjects were used for both hot and cold fluid test situations. They were treated as separate subjects in analyzing the results. In selecting subjects certain limitations were observed in addition to those mentioned above: 1. Patients who could not breathe with their mouths closed were rejected as subjects. 2. Patients who had ingested hot or cold food or fluids within the previous hour or who had smoked a cigarette within the previous 20 minutes were rejected. 3. Patients who had had strenuous physical exercise within 30 minutes of the test were rejected, since body temperature has been shown to return to a pre~exercise level within 25 minutes after exer- cise (DuBois, 1948). Seven subjects had engaged 5 in mild exercise consisting of sitting in a chair or walking for short distances from 15 to 60 minutes before the test. The remaining subjects were at bed- rest. Selle (1952) reported that body temperature begins to rise within an hour after and reaches and maintains a maximum three hours after eating. The length of time elapsed since food intake was recorded and ranged from 1.5 to 4.5 hours. One subject was in a fasting state. Date of onset of the last menstrual period was recorded, since body temperature has been shown to rise and remain above the previous level after ovulation (Selle, 1952). En- vironmental temperatures ranged from 72 to 89 degrees. A telethermometer was used to measure the temperatures. A telethermometer is an electronic device by which the temperature is read continuously on a dial at a point remote from the subject whose temperature is being measured. The temperature is measured by means of a probe with a temperature sensitive surface which is inserted into the subject and is connected to the telethermometer by an insulated wire. The particular telethermometer used was model 44 manufactured by the Yellow Springs Instrument Company, Inc. It is small and portable, weighing 5.5 pounds, and has a scale range of sensitivity of 68 to 108 degrees Fahrenheit (20 to 42 degrees Centigrade). Its power source is a flashlight 6 battery which the investigator changed weekly to prevent inaccuracies. The accuracy of the instrument is stated by the manufacturer to be ±l% of the scale range, or 0.4 degree of this 40 degrees scale. This model can accommodate 12 probe leads, facilitating the reading of comparative temperatures simultaneously by changing the channel selector switch. Two channels were used for the first part of the study when rectal temperatures were monitored. The oral probe is banjo shaped with the tempera- ture sensitive surface on the circular flat end of the probe. This surface is 1/16 inch thick and 13/32 inch in diameter. The tubular probe used for measuring rectal temperatures has a diameter of 5/32 inch and is 4.5 inches long. The standard measure of the speed of probe response is the time required to read 63 percent of a sudden temperature change to which the probe is exposed and is referred to as a time constant. Five time constants are required for a probe to read 99 percent of the total change. for the banjpprobe is 0.8 second. The time constant By four seconds 99 percent of the temperature change is recorded by this probe. The temperature reading 15 seconds after returning the probe to the mouth after the fluid was ingested should, therefore, be an accurate measurement of the mouth temperature. constants for the rectal probe are 17 seconds. Five time For the study this probe was not removed after insertion until test 7 completion, so only the original period of accommodation was of concern. In a pre-test situation the probes and a mercury thermometer of known accuracy were placed in a waterbath and read simultaneously. The telethermometer and mercury thermometer readings did not exactly correspond. A Pearson product moment correlation run on the two sets of temperatures yielded a coefficient of +.996. Points for the temperatures were plotted on a graph and were found to lie in a linear distribution. A regression analysis was done to determine the most appropriate site for a line through the points (Appendix A). The resulting line was used for converting the telethermometer readings obtained during the test into the clinically more meaningful mercury temperatures. Since the calibrations on the scale are in 0.5 degree increments, it was necessary to estimate smaller degree divisions. To standardize estimations, the investigator was the only person reading the temperatures and the telethermometer was kept in the same relative position to the eye level of the investigator. Hot and cold fluids were chosen as the two variables. In a pre-test on five subjects, it was determined that the average warmest temperature for drinking six ounces of hot fluid over 1.5 minutes was 133 degrees Fahrenheit (142, 137, 135,125 and 123 degrees). The temperatures of the 8 administered fluid ranged from 129 to 133 with a mean of 131 degrees. To prepare the fluid, water was heated, poured over instant chocolate in a styrofoam cup, stirred and allowed to cool to the desired temperature. Hot chocolate was selected in preference to coffee or tea to eliminate the variables of cream and sugar which could affect the fluid temperature. Iced water was prepared by adding four ounces of crushed ice to 2.5 ounces of cold tap water. This method achieved the desired temperature of 34 degrees Fahrenheit almost immediately and the temperature remained constant until all the ice had melted. Both hot and cold fluids were served in styrofoam cups and were taken through a straw over a 1.5 minute period. The styrofoam cups acted as insulators to help maintain the constancy of the fluid temperatures. In order to record the natural temperature variation, if any, over the test period, rectal temperatures were monitored in the first part of the study. However, when it appeared after 13 subjects that there were no significant changes due to the fluid ingestion and that the prospect of having a rectal temperature taken was limiting the number of patients agreeing to participate in the study, the rectal temperature monitoring was eliminated. The purpose and sequence of the test were explained to the subjects before obtaining their consent for participation 9 in the study. They were instructed not to talk, laugh, cough, yawn or to open their mouths during the test. It should be noted that plans for prevention of interruptions by phone were not made. Test times were scheduled so as not to coincide with visiting hours but an occasional visitor was encountered. Probes were inserted sublingually and rectally, and a baseline temperature was established by allowing for the period of accommodation required by the probes and by ascertaining the maintenance of the oral temperature at a constant level for three minutes. time. The fluid was prepared during this The oral probe was removed before and was reinserted immediately after the subject ingested the fluid. The oral temperature measurement was recorded every 15 seconds until the temperature reached and maintained a constant level for a 10 minute period. The rectal temperature was recorded only every minute since it was not expected to change in response to the local changes in the oral cavity and was not the area of prime concern for this study. To restrict talking to a minimum before the test, information regarding the time and contents of the last meal and the date of onset of the last menstrual period was recorded after completion of the test. CHAPTER III RESULTS The 15 subjects to whom hot fluid was administered are described in Table I by age, sex, response to the fluid in amount of change and time to return to the baseline, and the variations in oral and rectal baseline temperatures. Means of the ages, baseline temperatures, amount of change produced, the return to baseline time and the standard deviation of the return time are presented. The range of time required to return to the baseline temperature was 6.5 to 45 minutes with a mean of 15.8 ±10.6 (standard deviation) minutes. The range of change produced was 0.5 to 4.5 with a mean of 2.1 degrees. For the study a new baseline was considered established when a temperature was maintained for 10 or more minutes. It was decided that since an error of 0.2 degree in reading a clinical termometer could be made and that since medical diagnosis and treatment are not dependent upon minute variations in the temperature, a difference of 0.2 degree would be considered insignificant. On this basis six of the eight sub- jects who did establish new baselines had insignificant changes. It should be noted that the changes were enough to increase the group's mean temperature by 0.1 degree, from 11 TABLE I RESPONSE OF SUBJECTS DRINKING HOT FLUIDS Age Time to New Amount Baseline Return Baseline of Sex Temperature to Difference Change Degrees Baseline from Degrees Minutes Original Changes in Rectal Temperature 20 F 98.7 2.6 6.5 20 M 97.7 3.6 9.5 +.2 20 M 99.3 2.4 23.0 +.2 23 F 99.6 2.9 32.5 23 F 99.1 0.65 11.25 +.2 25 M 98.6 4.0 14.75 +.3 31 M 100.1 0.5 10.0 32 F 98.1 4.5 18.25 +.2 +.2 37 F 97.7 1.6 45.0 +.2 -.4 38 F 99.3 1.6 11.75 +.2 40 M 98.8 1.0 7.0 0 41 M 97.0 1.6 10.0 44 M 98.7 1.1 16.25 47 M 98.2 2.0 6.5 50 F 99.3 1.0 15.25 Means 32.7 98.7 15.8 2.1 ±10.6(S.D.) +.1 0 +.2 -.1 +.4 12 98.7 to 98.8 degrees. It is of interest that five of the eight subjects establishing new baseline temperatures were male, as were both of the subjects whose baselines varied by more than 0.2 degree. Unfortunately, rectal temperatures were not monitored on those two subjects, although no apparent correlation between oral and rectal baseline changes was seen in other sUbjects. To determine differences in response to hot fluid ingestion, subjects were divided into sex. categories at approximately the median of the age range. Table II presents the means calculated on these categories for comparison with each other and with the total group. Male and female subjects TABLE II MEANS OF AMOUNT OF CHANGE AND RETURN TO BASELINE FOR SUBJECTS DRINKING HOT FLUIDS IN SEX AND AGE CATEGORIES Group Number Mean Mean Combined of Change Return Mean Subjects Degrees Minutes Degrees Age Sex 20-34 F 4 2.7 17.1 20-34 M 4 2.6 14.25 35-50 F 3 1.4 24.0 35-47 M 4 1.4 9.9 20-50 F 7 2.1 20.1 20-47 ~-1 8 2.0 12.1 Younger Older Both Combined Mean Minutes 2.6 15.7 1.4 15.9 2.1 15.8 13 within an age group showed the same mean amount of change in temperature with the older age group having the least change. Male subjects were consistently faster in recovery of the baseline temperature. It should be noted that the female mean return time was influenced by the two extreme return times of 32.5 and 45 minutes. The longest return time for a male subject was 23 minutes. Because a 0.2 degree difference was considered to be insignificant, it is of interest to determine when the recovery process reached a point 0.2 degree from the terminal baseline level. This occurred at a mean time of 11.1 ±7.9{S~D.) minutes, or 4.7 minutes before the mean for full baseline recovery time. This corresponds well with the curve of the recovery process in which there is a constant decline until a point 0.2 degree from the baseline is reached, whereupon the curve develops a long flat tail (Figure 1). The responses of 16 subjects drinking cold fluid are presented in Table III. The range of time required for re- covery of the baseline temperature was 6.25 to 35 minutes with a mean of 16.6 ±8 (S.D.) minutes. The range of change produced was 1.2 to 8 with a mean of 5.1 degrees. Changes in oral baseline temperatures occurred in six subjects and were all greater than 0.2 degree. The changes produced a 0.1 degree decrease from the original mean baseline temperature of 98.4 degrees. Two subjects' 14 1..0 ('\1 'll) N ~ r M r-I N r-I H ~ 0 r-I ~ til Q) 00. +J :=$ £" .r-! ~ ~ 1..0 ll) ~ M N r-I o Baseline r-I 0 ~ 00 0 r-I ~ ~ r-I o Degrees Figure 1 The recovery process in 15 subjects qfter the ingestion of hot chocolate, plotted as the means of the recorded temperatures at one minute intervals. temperatures exceeded the baseline level. Both were female. In comparing the variables, no conclusions can be drawn; for one was at bedrest and one had had mild exercise; one rectal temperature did not change and the other dropped one degree; 15 TABLE III RESPONSE OF SUBJECTS DRINKING COLD FLUIDS Age Time to New Baseline Amount Return Baseline of Sex Temperature to Difference Change Degrees Baseline from Degrees Minutes Original Changes in Rectal Temperature 20 F 97.9 3.3 10.75 20 M 98.2 5.2 18.25 20 M 99.3 7.1 12.5 25 M 97.6 7.0 23.25 26 F 98.2 5.2 20.5 30 F 99.3 5.1 23.5 32 F 97.4 5.2 17.25 34 F 98.6 5.6 . 25. 75 35 M 98.9 3.6 7.75 35* M 97.7 3.1 14.25 40 M 98.6 4.8 11.25 44 F 97.7 5.5 35.0 +.5 47 M 98.6 4.4 11.25 -.4 47 M 98.6 7.2 6.25 -.4 -.2 50 F 99.3 8.0 6.75 -.4 -.3 52* F 98.2 1.2 21.75 Means 34.8 98.4 16.6 5.1 ±8.0(S.D.) *Partia1 lower denture removed +.4 0 -.5 -.3 -.2 -.7 in 35" -1. in 47" 16 one was 14 days into her menstrual cycle and the other had had three episodes of vaginal bleeding in the previous five weeks. The four subjects whose temperatures remained below the established baseline were at bedrest. Five of the six who established new baselines were in the 35 to 50 years age group. All rectal baseline temperatures dropped from the original level. Of interest is the observation that the two subjects whose oral temperatures dropped least had partial lower dentures which were removed during the test. Table IV presents means of amount of change and return times in sex and age categories for the subjects drinking TABLE IV MEANS OF AMOUNT OF CHANGE AND RETURN TO BASELINE FOR SUBJECTS DRINKING COLD FLUIDS IN SEX AND AGE CATEGORIES Group Age Sex 20-34 F Number Mean Mean Combined of Change Return Mean Subjects Degrees Minutes Degrees 5 4.9 18.0 Combined Mean Minutes 19.5 Younger 20-34 M 3 6.4 35-52 F 3 4.9 21.2 35-47 M 5 4.6 10.1 20-52 F 8 4.9 20.2 20-47 M 8 5.3 13.1 Older Both 5.5 19.0 4.7 14.7 5.1 16.6 17 cold fluid. Female subjects of both age groups had the same mean amount of change in temperature. The older group had the least mean change, as was seen with the subjects drinking hot fluids. The mean return times for female subjects in both hot and cold fluid groups are almost identical. The male subjects' mean return to baseline temperatures was approximately seven minutes faster than female subjects. The recovery curve after ingestion of cold fluid is similar to that after ingestion of hot fluid in that there is a steady increase until a point 0.3 degree from the baseline is reached. 2). The curve flattens out after this point (Figure The point at 0.2 degree from the baseline occurred at a mean of 14 ±6.9 (S.D.) minutes, or 2.6 minutes before the mean for full baseline recovery time. No observable influence on the recovery process by the environmental temperature, length of time since food ingestion, exercise or state of menstrual cycle was seen except in the instances cited above. Table V presents a comparison on the six subjects to whom fluids of both temperatures were administered to determine any similarities of response to hot and cold influences. Only one subject was consistent in response to both temperatures. Two subjects had return times lower than the group for both hot and cold fluid. 18 N N M N 0 N 0'\ M 00 M r--M \.0 M LO M o:::t4 M M M N M M M 0 U1 (J) 4J ::s s:: -r-! ~ M 0'\ 00 r--\.0 LO o:::t4 M N M 0 Baseline 0'\ 0'\ 00 0'\ r--0'\ \.0 0'\ LO 0'\ o:::t4 0'\ M 0'\ Degrees Figure 2 The recovery process in 16 subjects after the ingestion of iced water, plotted as the means of the recorded temperatures at one minute intervals. 19 TABLE V RETURN TIMES FOR SUBJECTS DRINKING BOTH HOT AND COLD FLUIDS Return to Baseline Cold Fluid Sex Return to Baseline Hot Fluid 20 F 6.5 10.75 32 F 18.25 17.25 20 M 9.25 18.25 20 M 23.0 12.5 25 M 14.75 23.25 47 M 6.5 11.25 Age Group mean for return to baseline for hot fluid: 15.8 Group mean for return to baseline for cold fluid: 16.6 CHAPTER IV DISCUSSION Implications for nursing practice can be drawn from the results of this study. Temperatures were found to be af- fected a significant amount by both hot and cold fluid ingestion. Oral temperatures taken immediately after these activities could result in recordings in which a fever state falsely appeared to be present or in which a fever state might be masked. By using the mean and standard deviation, it can be assumed that 84 percent (one standard deviation on the curve of normal distribution) of people drinking hot fluids like those used in the study will recover their baseline temperatures in 26.4 minutes. It was considered to be of more significance to determine when most people reach a point 0.2 degree from the baseline. Based on the mean and standard deviation, 98 percent (two standard deviations) can be assumed to have reached that point in 26.9 minutes. Similarly, with people drinking iced water, 98 percent can be assumed to reach 0.2 degree from the baseline in 27.8 minutes. The recommended length of time to wait to record an oral temperature after the patient has ingested hot or iced fluid is 30 minutes. This recommendation is based on the averages 21 of this group of subjects, and possible individual differences around this figure are to be anticipated. The use of a' telethermometer with a scale calibrated in smaller divisions would eliminate the necessity for estimations; one with a scale calibrated to 0.01 degree is manufactured. Use of a telethermometer equipped with a recorder for printing out the temperature curve would permit the investigator to look at other physiological responses to hot and cold fluid ingestion, such as pulse and blood pressure. It is suggested that this study be repeated with a larger sample to verify the results. Suggestions for further studies on oral temperatures with implications for nursing practice include: 1. Use of fluids at temperatures between the two extremes used in this study. 2. Investigation of febrile patients' responses to hot and cold fluid ingestion. 3. Investigation of responses of patients with fevers of different origins to hot and cold influences. 4. Addition of patients over 50 years to determine if age is a factor in amount and speed of response to fluids, especially cold fluids. 5. Exploration of the responses of people wearing dentures. REFERENCES Brim, Katherine and Chandler, Betty Alice. Changes in oral temperature. American Journal of Nursing, 1948, 48(12), 772-773. DuBois, E. F. Fever and the regulation of body temperature. Springfield, Illinois: Charles C. Thomas, 1948. Montag, Mildred D. and Filson, Margaret. Nursing arts. (2nd ed.) Philadelphia: W. B. Saunders Company, 1953. Nordmark, Madelyn T. and Rohweder, Anne W. Scientific foundations of nursing. Philadelphia: J. B. Lippincott, 1967. Price, Alice L. The art, science and spirit of nursing. Philadelphia: W. B. Saunders Company, 1965. Rasmussen, Sandra. Foundations of practical and vocational nursing. New York: Macmillan Company, 1967. Selle, W. A. Body temperature: its changes with environment, disease and therapy. Springfield, Illinois: Charles C. Thomas, 1952. Verhonick, Phyllis and Werley, Harriet. Experimentation in nursing practice in the army. Nursing Outlook, 1963, 11(3), 204-206. Woodman, Ellen, Parry, Susan M. and Simms, Lillian. Sources of unreliability in oral temperatures. Nursing Research, 1967, 16(3), 276-279. APPENDIX A 24 REGRESSION ANALYSIB LINE FOR ORAL PROBE. 106 105 104 103 102 OJ 101 ..Q 0 J..1 100 r-I 99 P-! co J..1 0 98 -4J 97 ."; OJ ~ s:: OJ 96 J..1 ~ co 95 ~ til OJ OJ 94 J..1 93 0 92 IJ'\ OJ 91 90 89 88 87 1 92 93 94· 95 96 97 98 99 100 101 102 103 104 Degrees Fahrenheit (Mercury Thermometer) 25 REGRESSION ANALYSIS LINE FOR RECTAL PROBE 105 104 103 102 101 Q) ..0 0 J...I p.. .--I ro 4J 100 99 98' 0 Q) p:: 97 4J 96 ~r-I ro ..c: s::Q) 95 J...I 94 ..c: ro f:t.I Ul Q) Q) S-.l 93 92 tp Q) Q 91 90 89 88 87 86 90 91 100 101 10 2 92 93 94 95 96 97 98 99 Degrees Fahrenheit (Mercury Thermometer) 03