A wealth of literature has accumulated which illustrates an association between salivary flow rate, salivary composition and oral health, but this link may be confounded if an individual’s flow rate varies from month to month. Previous investigations have shown that salivary flow rate fluctuates throughout the day, and that the unstimulated flow rate may be at its maximum in the mid-afternoon. Knowledge of this within-day variation allows its effect to be minimized by assessing salivary flow rate when it is at its least variable, for example in the mid-afternoon. Variations of salivary flow rates over longer intervals may also be present but have not so far been quantified in a temperate climate.

Shannon, I.L., Segreto, V.A., 1968. Periodic variations in flow rate and chemical constituents of human parotid fluid. Technical Report Sam-Tr. 1–14Shannon and Segreto (1968) investigated changes in sour-lemon drop-stimulated parotid salivary flow rate over a 1-month period. Their study was carried out in the subtropical climate of San Antonio, Texas, where 11 male naval recruits contributed a saliva sample every day. Changes were observed from week to week rather than from day to day. Later, Hyde (1972) investigated the paraffin wax-stimulated, whole salivary flow rate of children aged 5 and 6 years (n=48) in Vancouver, Canada; only children who had previous experience of chewing gum took part in the study. Hyde assessed salivary flow rates during July, August and the first 2 weeks of September before the children started kindergarten; subsequent to their start date, a repeat sample was collected during October, November and December. He observed that 22 of the 48 children had higher flow rates in the initial summer/autumn samples than the second, autumn/winter collection, and he attributed the decline in flow rates to the stress of starting school. In contrast, Crossner (1984), in Örebro, Sweden, collected paraffin-wax stimulated saliva samples at 10-week intervals on seven occasions (n=115); variation within participants accounted for only 16% of the total variation.

Variation in unstimulated whole-saliva flow rates over other time-spans has not been similarly investigated. In subtropical San Antonio, Texas, Shannon (1966) examined changes in the flow rate of unstimulated parotid saliva. New Air-Force recruits contributed a single sample at various times during 1 year; in all, 3868 saliva samples were collected. The participants who contributed to samples collected during the summer (from May until August) had a lower mean salivary flow rate (0.031 ml/min) than those tested at any other time during the year. The December and January groups recorded the maximum mean salivary flow rate (0.046 ml/min). Shannon suggested that this change over time was related to dehydration due to a combination of heavy physical exercise and climatic conditions.

Differences in results in these studies may be attributed to age, climate and the varying types of salivary flow rate measured. If variation in saliva samples exists throughout the year, this may have repercussions for salivary research. Our aim was to assess if unstimulated salivary flow-rate samples from children residing in a temperate climate varied over several months.

This research was carried out in Clwyd, North Wales, at two High Schools, one in Wrexham and the second in Flint, between September 1990 and June 1991. Approval was received from the South Clwyd Ethics Committee, North Wales before the study began. Forty-three participants aged 12 to 13 years at the beginning of the study were selected to take part; of these, 24 were male. This project was part of a larger study and 26 of the 43 children were also participating in a toothpaste clinical trial. Informed consent had been provided for all involvement in the two studies.

The study was timetabled so that samples would be collected at the beginning of each month; the study began on 6 September 1990. Sample collections from the same participant were never separated by more than 33 days. On only one occasion were sample collections less than 28 days apart; that is, between May and June, when they were separated by 23 days. A similar timetable of saliva collection was implemented in both schools where, for convenience, collections were staggered 1–6 days apart. Samples were collected in the afternoon at least 1 hr after eating or drinking. Collections were made between 13.15 and 14.45 hr Greenwich Mean Time for the months November until March. The other collections were made 1 hr later because British Summer Time was in force.

At each occasion, children provided a single, unstimulated saliva sample. No practice sample was collected before the first (September) collection, but the method was demonstrated at that time. The ‘draining’ method was used for salivary collection. Participants were instructed to swallow before sampling and if they needed to swallow during the collection they were advised to spit first, swallow, and then continue. A preweighed, plastic universal container (30 ml; Sterilin UK) was given to each child. These were labelled with their names and a unique identifying number. Participants were seated with their heads bent forward and saliva was collected for a timed 5 min. The bottle was then reweighed and the unstimulated flow rate (g/min) estimated. As the specific gravity of saliva is approximately 1.0, volume and weight were used interchangeably. Results are presented as ml/min.

Statistical analyses used the Statistical Package for Social Sciences, Version 7.0 (SPSS, Chicago, IL, U.S.A.). The variation of salivary flow rate within individuals from month to month was compared by repeated-measures ANOVA. When data were missing, as from among the total group’s (n=43) participants, missing values were estimated by averaging from the cases with a known value for that variable.

The attendance at each month ranged from a minimum of n=29 in February 1991 to a maximum of n=40 in May 1991. Eighteen children, eight males and ten females, attended all ten sessions; they are hereafter referred to as the “regular attendees”. Statistical analyses were carried out for the total group and the subgroup of regular attendees.

The consistency of individual measurements was assessed from the number of children remaining above the median and the number remaining below the median for each month. This analysis was confined to the regular attendees (n=18) for which the month-to-month measurements were directly comparable.

The ambient temperatures during the hours of collection were averaged (The Meterological Office, Berkshire, England). The 20 points represented by the school visits were compared, using Spearman’s correlation analysis, with the average temperature recordings during the hours of collection. The results took account of the hour adjustment necessary when British Summer Time was in force. These analyses were made on the total group (n=43) and the subgroup of regular attendees (n=18).

The September mean salivary flow rate (SD) in both the subgroup of regular attendees (n=18) and the total (n=43) was the lowest recorded throughout the study at 0.52 (0.30) and 0.49 (0.33). The highest mean flow rates were recorded in February at 0.87 (0.34) (n=18) and 0.80 (0.32) (n=43). There were no statistically significant relations between the variables age, sex and flow rate.

In the total group (n=43) the within-person variation, for the 9-month duration, was statistically significant (repeated-measures ANOVA; p<0.001). Significance was evident even when the September data (p=0.001) and the September and October data (p=0.02) were excluded.