U. S. Food and Drug Administration
Center for Food Safety and Applied Nutrition
FDA Prime Connection
NHDPHS - Outbreak of Y. enterocolitica With A Dairy Plant (11/95)
Report of the New Hampshire Division of Public
Health Services Bureau of Food
Protection on an outbreak of Yersinia
enterocolitica allegedly associated with a
New Hampshire dairy plant
November 1995
First Notification
The New Hampshire Division of Public Health Services (NHDPHS) was
first notified during the week of October 16, 1995 of a possible Yersinia
enterocolitica outbreak occurring in Vermont. The Bureau of Food
Protection (BFP) was notified of this by the Bureau of Disease Control as
yersiniosis can be a foodborne illness. On October 27, 1995 the BFP
received word that the outbreak was possibly linked to milk products from
a dairy plant located in New Hampshire.
Initial Discussion
A conference call was held on October 30 among the Vermont Department
of Health (VDH), the NHDPHS, and the Centers for Disease Control and
Prevention (CDC) in Atlanta. At this time there were 4 confirmed cases in
Vermont and one in New Hampshire. The 4 cases in Vermont had all
consumed milk from the New Hampshire dairy plant. The one case in
New Hampshire had not. Both the VDH and the NHDPHS invited CDC to
join the investigation. CDC said they would send Dr. Marta Ackers and
medical student John Markham to Burlington, VT the following day. CDC
also suggested we should hold off notifying the dairy that they were being
considered a potential cause for this outbreak. lt was suggsted that milk
samples be collected and tested for Yesinia. The BFP had collected routine
monthly finished product samples and a raw milk sample at the plant the
previous week on October 24, 1995. These would be tested as well.
U.S. Food and Drug Administration (FDA)
Stephen Pierson, northeast regional milk specialist of the Food and Drug
Administration (FDA) was notified by the NHDPHS BFP on October 31,
1995 of the Yersinia outbreak and its potential link to a dairy. The BFP
contacted the FDA to alert them of the possible link with the New
Hampshire milk plant and also to elicit information availabie on previous
Yersinia outbreaks. associated with dairy plants.
Sampling
Pasteurized milk samples were collected on October 31, 1995 by Ken
Mazur and Gary Quackenbush, Environmental Sanitarians with the
NHDPHS BFP. These samples were collected from grocery stores in
Keene, New Hampshire and Hanover, New Hampshire and delivered to
the New Hampshire Public Health Laboratories (NHPHL)on November 1,
1995. Four samples were collected by Ken Mazur. These included skim
milk code dated (11/6), 2% milk (11/6), whole milk (11/6) and chocolate
milk (11/3). Three samples were collected by Gary Quackenbush. These
included 2% milk (11/6), whole mitk (11/3), and chocolatc milk (11/1O).
The samples that had been collected by the BFP on 10/24 included skim
milk (11/3), 2% milk (11/3), whole milk (11/3), half & half (11/3), and
raw milk. These samples were all within standards when tested for
standard plate count, coliforms, butterfat, solids, phosphatase, and
antibiotics.
Contact with dairy
The first news the dairy plant received that they were being investigated in
relation to this outbreak was through rumor on the street. The BFP
received a telephone call from the dairy plant on Wednesday November 1,
1995. They wanted to know what was going on. Apparently someone had
heard someone else talking about the Yersinia outbreak and how it was
connected to the dairy plant. The rumor was confirmed and basic
information on the outbreak, i.e. how many people were affected was
given to the dairy.
Background on the dairy plant
The dairy plant is located on a dairy farm which is in the Connecticut
River valley in the western part of New Hampshire. It is a small family
owned and operated dairy. The family milks approximateiy 80 holsteins,
and there are approximateiy 80 more head of cows on the farm which are
not milked. There are also pigs, chickens and standardbred horses on the
farm. The horses and chickens are kept in separate barns. The pigs were
kept within the same barn but separated from the milking herd. Dry cows
are not housed near the pigs. Cows are miiked twice a day in a parlor.
The milk is piped from the parlor to a bulk milk tank in the farm milk
room where it is cooled and stored. The farm provides all of the milk
required by the processing plant. Approximately 1,1OO gallons (4,5OO
pounds) of surplus raw milk is sold each week to Agri-Mark, a milk
cooperative. The farm is part of the federally rated Agri-Mark BTU.
The processing plant was buiit in 1 992 and was first licensed in February
1993. The plant is composed of 5 rooms; the processing and filling room,
a walk-in cooler, a miik case storage and bottle washer room, a
multi-purpose room used to store single service caps, dispenser bags,
chocolate syrup, eggnog base, cleaning chemicals, milk cases, and a
recirculated cooling water storage tank, and a small office where milk is
sold from the plant. The multi-purpose room is also used as a dressing
room for employees who are going from the barns into the plant. All
rooms are adequately sized. All rooms are separated from each other by
solid partitions. The doors in these partitions are solid and self closing.
There are no doors between the processing room and the farm milkhouse.
There is an opening in the wall between the bottle washer and the filler
adequate for the bottle track and bottles to pass through the wall. All
floors are of smooth finished concrete and are sloped to trapped drains.
Walls and ceilings are constructed of white fiberglass reinforced plastic
board. Toilet and handwash facilities are located off the office area and
are separated from the plant by two solid doors. It is plant policy to not
allow workers to enter the plant from the barns without changing their
clothes and their boots. A diagram of the dairy barn, milking facility, and
processing plant are included in appendix A.
The plant is in operation on Monday and Thursday of each week.
Approximately 2,300 gallons (11,000 pounds) of milk is pasteurized and
bottled on each of these days. The milk is piped through the wall of the
milk room into the processing room. The plant produces the following
milk products: skim milk, 2% lowfat milk, whole milk, chocolate milk,
half & half, and heavy cream. Eggnog is also processed in season. All of
the products are packaged in returnable glass bottles; 1/2 liter, quart and
half gallon sizes, or in 5 gallon dispenser bags. There are four people who
work in the plant when it is in operation. There are also 2 drivers who
deliver the milk. Deliveries are made in 2 mechanically refrigerated trucks
on Monday and Thursday aftenoons after processing and on Tuesdays and
Fridays. The delivery trucks, which are owned by the dairy plant, are
loaded after processing and bottling are completed for the day. The dairy
distributes all of its own milk. The milk is distributed in a 20 mile radius
of the plant and there is also a deiivery route down Route 91 to
Brattleboro, Vermont and Keene, New Hampshire. Approximately 20% of
the milk products are distributed in Vermont. The plant is not on the
Interstate Milk Shippers list as Vermont does not require plants shipping
milk into the state to be on the list. This plant is however, inspected
according to the Federal Pasteurized Milk Ordinance (PMO). Inspections
are made and equipment is tested every three months as required by the
PMO. Milk samples are collected four out of every 6 months as required
by the PMO.
The water supply for the plant consists of two 1.000 foot drilled wells
located over 1/4 mile from the plant. Water from the wells is commingled
in a buried, sealed concrete reservoir located in front of the plant. The
water was tested for coliform on October 24, 1995 and had a result of
5.1. Water is heated for plant use in a boiler. Milk product cooling water
is from a recirculated system. The recirculated cooling water had a
coliform result of 1.1 when tested on October 24, 1995.
Dairy Plant and Farm Inspection on November 2, 1995
On November 2, 1995 Leah Keller, Dairy Sanitation Supervisor, Chuck
Metcalf, Environmental Sanitarian and Joyce Welch, Administrator with
the NHDPHS BFP visited the farm and pasteurization plant. This was a
processing day. The purpose of this visit included:
1. Talking with the family to explain that there was some indication
that people who were ill with Yersinia may have consumed milk
from their dairy. At this time, CDC had not conducted the case
control study so there was no statistical link.
2. Discussion about Yersinia and how it is transmitted.
3. Observation of the pasteurization process.
4. General observation of the plant and farm.
Upon arrival at 11 am it was noted that the plant was in operation. After a
brief discussion with the plant owners, BFP inspectors observed the bulk
tank room, milking parlor, and barn and then proceeded inside the plant.
The milk room and parlor are adequate for their use. The milk room
contains a large bulk tank which is used for raw milk storage, a smaller,
unused bulk tank, wash sinks, a hand sink, the boiler, the compressor for
the large bulk tank, a desk and filing cabinet. The parlor is a double 4
with two entrance doors and one exit door for the cows. No major
changes from previous inspections were noted in either the milk room or
the parlor. The freestall where the cows are housed was clean and dry.
The cows were also clean.
The plant was clean and well lit. There was adequate shielding over the
bottles between the bottle washer and the filler and between the filler and
the capper. There appeared to be adequate ventilation. There was one
person engaged in processing the milk, one person operating the bottle
washer, and two people running the filler, running the bag-in-box filler as
needed, and moving product into the cooler. Filled bottles are placed in
plastic milk cases which are then placed in the cooler. Clean clothing and
boots were worn by the employees. Hair restraints were not worn. There
was no sign of anyone eating or drinking in the plant.
At the time of observation, half & half was being processed. This was
foliowed by a water recirculation and then heavy cream. The recording
thermometer read 176xF. All regulatory seals were in place. The
separator was torn down for washing. The cones were washed in a clean
-out-of-place (c-o-p) tank. The rest of thes eparator was washed by hand.
It was noted that there was no place other than the floor to wash and
reassemble some parts of the separator. At one point the person engaged
in processing did go into the milk room and return to the plant. No foot
baths were in place. Foot baths filled with a sanitizing solution and located
at doors into the plant were recommended for washing footwear to prevent
cross contamination.
The bottle washer is located in a separate room. Bottles are fed in through
the front, go through a caustic soak tank, are rinsed and returned out the
front. Bottles are automatically placed on the track. Excess rinse water
from the bottlewasher is used to rinse off milk cases as they are emptied.
There is no caustic or sanitizer in this water. The process of rinsing the
cases is accomplished by dumping a 5 gallon bucket of the excess rinse
water over a stack of cases. The cases are then manually pushed across
the floor from the bottle washing room into the processing room to be
used. This method of rinsing the cases was identified as a poor practice.
Because Yersinia is frequently found in pigs, special attention was paid to
where the pigs were located. Nine pigs were observed in stalls in the
northern section of the cow barn attached to the plant. Cows were located
in the southern section of the same barn. There is approximately a 50 foot
separation between the cow freestall and the pig pens with a barricade
between the two sections. Although the cows and pigs were under the
same roof, they were not in close proximity to each other. The pigs were
clean and bedded in clean sawdust. The troughs were clean and dry. Care
and feeding of the pigs does not occur while the plant is in operation. The
person responsible for the care and feeding of the pigs is not directly
engaged in milk bottling operations when the plant is in operation. There
appeared to be no physical cross connection between the pigs and cows.
A sample of well water was taken from a hose bibb near the bottle washer
during this visit for analysis for Yersinia. Subsequent to the above
observations, there was a final discussion with the plant owners. The
possibility of cross contamination between the barn and the plant was
discussed. The two areas of possible cross contamination are: (1) the back
room where the sweetwater tank is located and where employees remove
their coveralls and boots, and (2) the milk room. Foot baths were
recommended. Concem about the case washing practices was also
expressed. Other information collected during this conversation included
the following:
Products: The following product sizes are produced at the plant:
Whole milk (1/2 gals qt.); 2% (1/2 gals qt.); skim (1/2 gals & qt.);
chocolate milk (l/2 gals & 1/2 L); 5 gal bags in whole, 2% skim,
and chocolate; cream (l/2 gals 1/2 L); half & half (qt. & 1/2 L) No
other items are produced on the farm at this time.
Employees: It was asked if any of the farm workers had been ill
with diarrhea recently, and the response was no.
Livestock: Six of the nine pigs were bought on August 2l, 1995.
No other animals are purchased.
Illness in domestic or farm animals: When asked if any animals
(including domestic pets) had been sick recently, the response was
that only one cow had been sick the previous weekend. This animal
was sick for one day.
Production of chocolate milk: Chocolate and raw milk are mixed
before pasteurization. The mix is blended for 15 minutes before
pasteurization.
Production procedures: It was asked if any production procedures
had changed recently. The answer no. It was mentioned that a
milk pump had ceased operating in the barn several months before.
This had held up milking for about 6 hours.
Distribution of 5 gallon bags: A list of establishments receiving 5
gallon bags was obtained. This was telephoned to the NHDPHS
BDC upon leaving the plant.
Groups touring the farm: During September and October, 1995,
only two groups toured the farm. In September a group of Danish
tourists were brought to the plant by Steve Taylor, the
Commissioner of the New Hampshire Department of Agriculture.
These were mostly older people. On October 18, 1995 two sessions
from the Hartland Kindergarten toured the plant. The plant was
not operating this day. Thc children were given chocolate milk to
drink. There were other groups scheduled for tours, but these
tours were canceled.
Returned milk: Returned (out-of-date) milk is taken directly to the
back room where the sweetwater tank is. Milk is poured from the
bottles into white plastic buckets which are used to carry the miik
to the pigs. The white plastic buckets are washed outside and
stored in the back room. These buckets never enter the processing
room. On occasion when the weather is hot and the milk may not
be fed to the pigs in a timely manner, a case or two of the returned
milk may be stored in the cooler so the milk does not spoil.
Bottle washing procedure: Bottles are soaked in caustic solution at
approximately 140xF, and then subjected to 3 rinses (beginning
rinse 120x-130xF, next rinse 80xF, final rinse clear water).
Cooler temperature: 36xF.
Upon arriving back at Concord there was another conference call between
New Hampshire and Vermont. Also included in this call was Dr. Clifford
McGinnis, State Veterinarian with the New Hampshire Department of
Agriculture (NHDA). During this call it was decided that M. Geoffrey
Smith, M.D. M.P.H., NHDPHS State Epidemiologist, Leah Keller, BFP,
Marta Ackers, CDC, and John Markham, CDC, would go to the farm on
the following day to talk with the family again. The family would be
requested to obtain stool samples from themselves and the plant workers
to be tested for Yersinia. There was discussion of culturing the pigs for
Yersinia, however it was later dccided that this would not be done at this
time.
Dairy Plant and Farm Visit on November 3, 1995
M. Geoffrey Smith, NHDPHS BDC, Leah Keller, NHDPHS BFP, Marta
Ackers, CDC, and John Markham, CDC, met with the dairy's owners at
the dairy at 9:00 a.m. and then proceeded to a family member's house to
discuss the situation. About 1 1/2 hours were spent talking with the
family. Dr. Smith opened the talk by discussing Yersinia, symptoms, etc.
The family had numerous questions to ask Marta Ackers regarding her
investigation. The family expressed unhappiness with the fact that no one
was there from the VDH as most of the confirmed cases were from
Vermont.
Thc family was again requestcd to provide stool samples to be tested. Dr.
Smith reviewed the written procedure for taking stool samples and left the
stool kits with the family members.
After the discussion, the invtigating team was given a tour of the farm.
The processing room was viewed from the front office and the bottle
washing room but no one actually entered that room. The bottlewasher,
cooler and back room were also observed. The tour then went in through
the north end of the barn to view the pigs and the cows. Everything
appeared much as it did the day before. The pigs were clean, their pen
was clean and the troughs were dry. The cows were also clean and the
freestall appeared well maintained.
The farm tour was concluded and the group left the farm at approximately
11:45.
Conference Calls
Subsequent to the two farm visits, conference calls were held on
November 7 and 9, 1995. The call on November 7 included the VDH, the
NHDPHS, Dr. Clifford McGinns and Dr. Joseph Moore, Director of the
Veterinary Diagnostics Laboratory at the University of New Hampshire.
This call was mostly a review of the known facts of the outbreak.
The conference call on November 9 included VDH, NHDPHS, CDC,
FDA Washington, and the FDA regional office in Stoneham. At this call it
was decided that a team should go to the plant on the following day. The
purpose of this visit would be to get as much information as possible on
the farm, collect environmental sampies, culture the pigs, do a thorough
break down of the equipment, swab tanks and drains, etc.
Farm Inspection on November 10, 1995
The team assembled at the farm consisted of Dr. Smith, NHDPHS, Leah
Keller, NHDPHS, Dr. Marta Ackers, CDC, Stephen Pierson, FDA, Dr.
Clifford McGinnis, NHDA, Dr. David Kluesner, USDA, Lester Colby,
NHDA.
The team arrived at the dairy around 11:00 am. The owners and a number
of other famiiy members were at the dairy. Afterinitiai introductions and
discussion, a review of the dairy began. Dr. McGinnis, Dr. Kluesner, and
Lester Colby collected tongue, throat and rectal swabs from the 9 pigs.
Thc other team members were involved with the inspection of the
pasteurization plant.
Milk processing equipment inciuded: a high temperature short time
(HTST) pasteurizer with a single stem flow diversion valve (FDV), a
homogenizer which is used as the timing pump, a c-o-p tank, a 150 gallon
raw blending tank, 3 pasteurized product tanks: a 200 gallon tank, a 220
gallon tank, and a 500 gallon tank, 2 filling machines; a 6 valve rotary
glass bottle filler and a dispenser bag filler, a miik separator, a raw
positive displacement pump, a raw centrifugal pump, a pasteurized
centrifical pump, 10 gallon cream cans, and a soaker type bottle washer.
All milk contact surfaces are constructed of stainless steel with food grade
rubber gaskets used as needed.
The following information was provided by the dairy plant owner:
Processing days begin at 7:00 am. The process begins with tne owner
starting up the pasteurizer. It is brought up to 190xF and recirculated for
about half an hour to hot water sanitize the unit. Cut-in and cut-out
temperatures are tested by the operator when the system is being heated.
Iodine sanitizer is mixed in the c-o-p tank and circulated through the raw
lines.
Iodine is used to sanitize the raw and pasteurized milk storage tanks. This
is done by using a water hose that is colmected to an iodine injection
system to make a sanitizing spray. Sanitizer is coliected in the pasteurized
product tanks and then pumpcd to the filler and the bag-in-box filler. The
sanitizer stays in the filler bowl for about 1 hour. Sanitizer strength is
determined by a visual check and by following label directions for the
bottle of iodine in the injection system. Test strips are used occasionally.
Iodine sanitizer is used for the cream cans. The separator is not sanitized.
The PMO requires that all product contact surfaces be sanitized before
each use. However, the separator is used only for raw milk prior to
pasteurization which would effectively kill Yersinia. After the pasteurizer
has been hot water sanitized, the hot water is run through the pasteurized
lines to the pasteurized product tanks to sanitize the lines. This is after
the pasteurized product tanks have been sanitized with iodine.
Around 7:30 a.m.- 7:45 a.m., the owner starts the separator. Full cans of
raw cream go into the cooler. There are usually 6 - 7 cans. Enough milk
is separated to make both skim and 2% milk. The 2% milk is
standardized after pasteurization by adding pasteurized whole milk to
pasteurized skim milk. Pasteurization begins around 8:00 a.m.
Pasteurized skim milk for bottling goes into the 200 gallon round tank.
Skim milk is always bottled out of this tank. Skim milk next goes into the
220 gailon square tank and finally into the 500 gallon tank. The 220 and
500 gailon tanks are used for 2% milk. When these two tanks are filled
to the proper marks for making 2%, more skim milk is added to the 200
gallon tank while it is being drawn out to the filler.
After skim milk pasteurization, bottling, which begins around 9:00 a.m.,
and processing occur at the same time. Whoie milk is stored in the 200
gallon tank after it is empty of skim milk. Whole milk may aiso be stored
in the 220 gallon tank if more roon is needed. Half & half is stored in the
500 gallon tank. Cream is stored in thc 200 or 220 gallon tank. Chocolate
miik is stored in the 200 gallon tank. If any tank is left empty for more
than a short time, it is sprayed with a sanitizing rinse and drained before
more product is put in the tank.
The separator is broken down for washing while whole milk is being
processed. Also while whole milk is being processed, the raw storage tank
is sanitized and half & half is blended in it prior to pasteurization.
Chocolate milk is mixed in the raw tank afer the half & half has been
pasteurized. Cream is poured directly from the cream cans into the
constant level tank of the HTST.
Dispenser bags are periodically filled during the bottling period. A three
way valve is used to adjust the flow of miik to both filling machines when
the dispenser bags are filled. The product flow rate to the bag filler is
slow enough to allow one operator to move between that filler and moving
bottled product into the cooler. The filler operator controls the start/stop
of the bottle washer and manually loads filied bottles into cases. A quality
assurance pull date is mechanically applied to the pogs on all bottles after
capping. This pull date is 11 days from the date of processing.
Prior to filling, a Continental caustic soaker type bottle washer is brought
up to operational temperature. Indicating dial temperature gauges for the
soaker section, and the first two rinses are located on the washer unit
where the washer operator can see them. Temperatures are not required to
be recorded. Sodium hydroxide (NaOH) is used to chemically treat the
soaker section of the washer for cleaning and bactericidal treatment of the
bottles. The caustic solution strength is determined daily by the operator
and NaOH is added as needed. Records of caustic addition are maintained
and were observed. Records of daily caustic strength are also maintained
and were observed. Caustic strength is maintained at 3.5 to 3.7% NaOH.
The bottles remain in the caustic for approximately 6 minutes at 140xF.
The bottles are rinsed in 3 stages; a 120xF caustic water rinse, an 80xF
clear water rinse, and a 45xF clear water rinse. The final two clear water
rinses are with untreated water. The soaker section was last drained in
September 1994. There are no requirements for draining bottle soakers
and subsequent examination of the metal cups used to hold the bottles did
not show any soil build-up which would indicate the soaker section was
not clean.
The 120xF soiution tank is drained bi-weekly and the 80xF rinse tank is
drained daily. The 45xF water is drained as it is used. It was noted that
the door between the bottle washing area and the processing plant is not
tight fitting.
Processing is generally completed by 1:00 p.m. The pasteurizer is rinsed
to flush out remaining product at the end of processing. This liquid is
discharged into a separate holding tank and is spread on the fields. Clean
up of bottling equipment and the plant is generally completed around 3:30
p.m.
Finished product is stored in the walk-in cooler. The cooler is equipped
with mechanical cooling and was 36xF. Product is placed in the cooler in
a counter clockwise pattern starting at the product load out end. Product
stored under or adjacent to the refrigeration unit are half & half and
cream. The condensate line terminated above product and was dripping
onto the product. The cooler was otherwise clean. At the time of this
inspection, the cooler did not appear to have an area designated for
returned product.
Deliveries are made using 2 mechanically refrigerated trucks. Returned
milk and glass bottles are collected by the delivery personnel and returned
to the plant. Loading and unloading occur at the same truck dock area.
The truck dock opens into a small staging area. This staging area is
contiguous with the bottle washing area with no door separating the two
areas. The cooler also has a door opening into this staging area. Clean
cream cans were observed in this area as well. Empty cases and returned
product are received in the plant between 3 and 5:00 p.m. on delivery
days.
Environmental swabs to be tested for the presence of Yersinia were taken
by Leah Keller, NHDPHS BFP from the following locations:
1. Raw milk positive pressure pump.
2. PVC connection on water hose by HTST.
3. Water hose by balance tank.
4. Inside of 500 gal tank - underside by agitator.
5. Pasteurized centrifugal pump.
6. Underside of glass bottle filler.
7. Condensate in upper corner by cooler and bottle washer doors.
8. Exposed insulation on steam line above 220 gal square tank.
9. Floor drain by glass bottle filler.
10. Back of soaker.
11. Final rinse area of bottle washer.
12. Wall in back storage room where buckets for returns are stored.
13. White water drain behind bag-in-box filler.
14. Floor drain in cooler.
15. Compressor condensate line in cooler.
16. Floor drain by bottle washer.
The locations for the swabs were based on where milk or water containing
Yersinia might collect undetected and where post pasteurization
contamination of bottled milk might possibly occur. These locations were
chosen by Stephen Pierson.
Conference Call on November 14, 1995
A conference call was held on November 14, 1995. In addition to
NHDPHS, VDH, CDC, FDA, Washington, and FDA, Northeast Regional
Office, Stoneham, this call also included Vermont Department of
Agriculture, and the FDA office in Augusta, Maine. This call discussed
the ongoing investigation. As a resuit of this call, it was decided that an in
depth examination of the dairy needed to be conducted. This would
include a visit on a non-processing day to break down and test equipment
and a visit on a processing day to observe all processing operations from
start-up in the morning through the end of bottling.
Dairy Plant and Farm Inspection on November 115, 1995
The team assembled consisted of Dr. Smith, Aime Ramsey, NHDPHS
BFP, William Bolton, NHDPHS PHL, Leah Keller, Marta Ackers,
Stephen Pierson, Byron Moyer, VTDA. The team arrived at the farm
around 1:00 p.m.
Stephen Pierson, Aime Ramsey, William Bolton, and Leah Keller
observed as the plate heat exchange portion of the HTST was
disassembled and opened for inspection. The cooling, milk regeneration,
and heating sections were visually examined. No structural or cleaning
abnormalities were observed. Drain holes were observed in two raw plates
in the milk regenerator section. The flow diversion valve, holding tube,
and vacuum breaker were also disassembled. No abnormalities were noted
in these areas. One valve of the homogenizer was also partially
disassembled. There were no abnormalities there. The homogenizer had a
regulatory seal attached to the cover over the gearing mechanisms. The
Taylor recorder-controller had a regulatory seal over attached to the cover
over the cut-in/cut-out controls.
The FDV micro switch was evaluated by removing the valve stem key and
also by turning the hex nut. The micro switch functioned properly for both
tests. There was no leakage observed past the valve seat. Diverted flow
holding time was checked and found to be 22.8 seconds on two tests.
Overall, no deficiencies were noted on the plant equipment.
Pasteurizer charts maintained by the plant were reviewed by Byron
Moyer. Charts from 8/3/95 to 11/13/95 were reviewed. Two charts
recorded unusual occurrences: a boiler failure on 8/31/95 and the
homogenizer valve seals were replaced on 9/18/95. Two anomalies were
noted: the cut-in and cut-out temperature points shifted from 163/162.5 on
9/21/95 to 164/162 on 9/25/95, and the frequency pen was not recording
in the same arc as the temperature pen from 8/3/95 to 8/10/95 and from
9/4/95 to 9/21/95. Products pasteurized during these periods were noted to
be pasteurized at above cut-in temperatures, and there were no FDV
diverts on product during these periods. It was later determined that the
change in cut-in and cut-out temperatures was made when a service
technician adjusted the HTST recorder-controller on 9/24/95. A regulatory
signature was noted on the chart dated 9/18/95. The charts indicated that
miik is pasteurized in the following sequence: skim, skim milk for 2%
whole milk for 2%, whole milk, half & half, cream, and chocolate milk.
Skim, 2% lowfat milk products, and whole milk are pasteurized at 170xF.
Higher fat and flavored products are pasteurized at 175xF.
Milking procedures were evaluated by Stephen Pierson. One operator
cleans the udders and pre-dips the teats with an iodine teat dip soiution.
This solution is wiped off prior to attaching the miiking machines. The
operator wears coveralls over personal clothing.
The delivery trucks were also evaluated. They are equipped with soiid,
tight fitting doors with strip curtains. The trucks appeared to be clean.
Environmental swabs to be tested for the presence of Yersinia were taken
by Wiliiam Bolton, NHDPHS PHL, from the following locations:
1. Front wall and floor area of Mac delivery truck.
2. Side door and plastic strip curtains of Mac truck.
3. Rear door and floor of Mac truck.
4. Front wall and floor area of International delivery truck.
5. Side door and plastic strip curtains of International truck.
6. Rear door and tloor of International truck.
7. Cream cans.
8. Floor beneath crates cream cans are stored on.
9. Returned bottle cases.
10. Floor beneath returned bottle cases.
As with the swabs taken on November 10, 1995, these locations were
chosen for their potential for collecting milk or water that may contain
Yersinia and for their potentiai for contaminating bottled milk. These
locations were chosen by William Bolton with direction from the previous
conference call and input from Stephen Pierson and Steven Simms, FDA.
A sweetwater sample was also taken to be tested for Yersinia.
Dairy Plant and Farm lnspection on November 16, 1995
Team members assembled at the farm included: Byron Moyer, Leah
Keller, Stephen Pierson, and William Bolton. Observation of the
processing day began at 7:00 a.m.
This review confirmed that the pasteurizer is sanitized with hot water
(182xF) for about one hour. Iodine is used to sanitize raw lines and
pasteurized surge tanks. Water used to sanitize pasteurized lines was
160xF where it discharges to the floor. This is not where it discharges
from the lines being sanitized and the water loses heat from that point as it
travels through the pasteurized tanks and some piping before it drains on
the floor. One pasteurized line was sanitized for about 5 minutes. The
other lines were sanitized for less than 5 minutes. The lines are required
to be exposed to 170xF for 5 minutes to be adequately sanitized using hot
water. The operator also neglected to sanitize the cap to the filler cover.
When this was pointed out, it was done. According to the operator, this is
routinely done. After the filler is emptied, the rubber fill heads are placed
in an iodine solution before being replaced on the filler. It was noted that
the 200 gallon tank was re-sanitized between whole milk and chocolate
milk. The filler bowl was also given a sanitizing rinse between cream and
chocoiate milk.
Prior to separation, vitamins are added to the raw milk in the milk house
tank with a syringe. A vitamin A & D concentrate is used. Label
directions state that the vitamins should be added after separation.
However, the vitamin concentrate is water soluble so the vitamins would
tend to go with the skim milk when separated and would end up in the
skim and lowfat products as required. Raw milk then goes through a
stainless steel pipeline from the milkhouse the plant where it is separated.
There were no pipeline circuits that created a cross connection betwcen
raw milk and pasteurized product. Cream coming off the separator was
38xF. The temperature of cream in a full can when placed in the cooler
was 42xF. All cream separated was used during processing on this day.
Water was recirculated in the HTST between half & half and cream and
between cream and chocolate milk. The indicating thermometer read
170xF and the recording thermometer read 170xF at 10:10 am. The
indicating thermometer was 176.5xF and the recording thermometer was
176.5xF at 11:15 am. Manual divert was checked and marked on the
chart. It was noted that the hose used to run the final rinse of the
pasteurizer before washing down the drain to a holding tank is long
enough to reach the drain but not long enough to be placed down inside
the drain.
The temperature in the bottle washer soaker section was found to be
140xF which was consistent with the thermometer reading on the bottle
washer. The bottles remained in the caustic solution of 3.6 for more than
6 minutes. The caustic solution is more than 2 times stronger than what is
required for that time and temperature by the National Soft Drink
Association and the Pasteurized Milk Ordinance (PMO). The steel cups
which hold the bottles in the washer were free of scale and other soil build
ups. Ventilation in the plant seemed to move from the processing area out
to the bottle wash room.
Cases for dispenser bags were scrubbed with soap and water in the bottle
wash room before the bottle washer was startcd up for the day. Cases for
bottled milk were not washed on this day.
The dispenser bag filler is not equipped with a cap removal and
attachment device. Dispenser bag caps are manually removed and placed
into a stainless steel basin containing an iodine sanitizing solution. Caps
are then removed from this solution and replaced on the bags when they
are full.
Foot baths were in place inside the door from the office and also inside
the door from the bottle wash room. The iodine solution in these baths
was changed several times during the day. There were also several iodine
hand dip stations, and buckets containing iodine solution where the wheels
which move the different sized bottles through the filler were kept when
not in use. Two employees wore hair restraints, two did not.
Tubes containing cap stock were placed on a stainless steel table for
staging. This table had a sag in it which had accumulated water. The
outside of the tubes of cap stock were noted to be wet.
The following items were noted and corrections outlined:
1. The gasket on the outlet valve of the 22O galion tank was in
poor repair. It was not soiled. The plant should establish a
routine for checking and replacing gaskets.
2. Vitamins are added prior to separation. The plant was
instructed to contact the vitamin manufacturer to make sure
that this practice is acceptable.
3. It could not be ascertained that pasteurized lines received
the proper time and temperature requirements for hot water
sanitization. The plant was advised to loop the lines for the
hot water sanitize so that it is incorporated with the
sanitization of the pasteurizer.
4. Several separator parts are washed on the floor. The plant
was instructed to provide rubber floor mats which the
separator could be washed on.
5. The tubes of cap stock were exposed to moisture on the
staging table. The plant was instructed to drill another
drain hole in the table.
6. The plant was reminded to sanitize the small cover on the
filler bowl.
7. It was recommmded that when the dispenser bag covers are
removed from the sanitizer solution before being placed on
the bags, they be given a shake to remove excess sanitizer.
The pick up schedule for raw milk by Agri-Mark was discussed.
Agri-Mark picks up surplus milk on Thursday p.m. one week and Friday
p.m. the next week. The Thursday pick up empties the farm tank. The
Friday pick up may only partially empty the farm tank. The tank is
emptied and cleaned each processing day. If the tank is not emptied on
Friday, it may go from the Thursday evening milking to the Monday
processing without being emptied and washed. When the tank is emptied
on Thursday p.m. it goes from the Friday morning milking to Monday
processing without being washed. During the week the farm bulk tank
goes from the Monday evening milking to Thursday processing without
being washed.
The wells were also examined. One of the well head caps was not bolted
down, however, this well head was covered by a steel drum. There was
animal manure spreading equipment observed stored near this well. There
is also a man made surface water run off collection pond observed up
grade and over 200 feet from this well. Each well produces 2 - 3 gallons
of water per minute.
Samples of all product produced on this day were collected for culture for
Yersinia.
Review of Regulatory Records
A review of the New Hampshire Public Health Dairy Sanitation Program's
records on the dairy was conducted on November 17, 1995 by Stephen
Pierson, FDA. This review covered pasteurization test rccords, raw and
finished product sampling records, potable and recirculated cooling water
test results, and inspection reports.
The records for water testing showed that the sweetwater had tested
positive for coliform on 10/24/95. It tested negative on 11/14/95. The
well water was contaminated with coliformson both dates. A subsequent
test on 12/19/95 was negative for coliforms.
The plant had last been inspected on 9/19/95, a non-processing day. The
following items were debited: doors not self closing, pipeline hole into
milk house not covered, fan louvers not closed, rough weld on raw
storage tank agitator shaft, soaker caustic records not maintained, cabinets
for single service articles not closed, dispenser bags not in cabinet,
frequency pen out of reference arc. It was also noted that shielding was
not in place over bottle conveyor.
Raw and finished product samples had been collected to meet 4 out of 6
month requirements. Results for samples collected since July are included
in Appendix B.
The pasteurizer was last tested on 9/29/95 after the seal on the
recorder-controller had been broken by a service technician for routine
cleaning and maintenance. A complete equipment test was conducted at
this time. The previous equipment test had been conducted on 7/18/95.
The pasteurizer had been timed during that test. The micro switch wire
was broken at that time and was repaired.
Sample Results
Samples were collected on the foliowing dates for Yersinia testing:
~ Finished product: 10/24/95, 10/31/95, 11/14/95 and
11/16/95
~ Raw milk: 10/24 and 11/14
~ Well water: 11/2
~ Sweetwater: 11/15
~ Environmental swabs: 11/10 and 11/15
~ The pigs: 11/10
Lab results are included as Appendix C.
Noted Deficiencies and Corrections Made By The Plant
1. Improper case washing. Case washing has been
discontinued. A case washer has been located and will be
installed.
2. Water supply is contaminated with coliform bacteria. The
wells have been chlorinated. If problems persist the BFP
will consult the Department of Environmental Services.
3. Final rinse from bottle washer is with untreated well water.
An iodine injection system was added to treat the water for
the final rinse on 12/12/95.
4. The condensate line from the compressor in the cooler was
dripping on product. This line was repaired on 11/14/95.
5. Two of the pasteurized product lines wcre not properly
sanitized at 170xF for 5 minutes. The plant will closely
monitor the sanitizing of these lines to assure that they are
subjected to 170xF water for a minimum of 5 minutes.
6. Employees do not wear effective hair restraints. The
employees will wear caps.
7. Tubes of caps on table with accumulated water. A new hole
has been drilled in the table to drain any water that may
accumulate.
Additional steps taken by the plant include:
1. Foot baths have been installed.
2. The pigs have been slaughtcred. even though samples from
the pigs tested by NHDPHS PHL were negative for
Yersinia.
3. All items from the previous plant inspection have been
corrected.
Additional Surveillance By NHDPHS BFP
The BFP Dairy Sanitation Section will provide additional surveillance of
the plant by:
1. Conducting monthly inspections of plant conditions and
procedures for the next six months.
2. Conducting an equipment test in December and every three
months thereafter.
3. Collecting samples of finished product and raw milk from
the plant on a monthly ongoing basis to be tested for
coliforms, standard plate count, and Yersinia.
4. Collecting well water and sweetwater samples on a monthly
basis for the next six months to be tested for coiiforms and
Yersinia.
Commentary
The coliform count in the well water is of some concern. When a water
source has a positive resutt for coliforms, it is the policy of the BFP to
encourage the owner of the source to have it disinfected. Another sample
is generally taken within thirty days if there has been adequate time for
any added disinfectant to have cleared fron the source. If that test also has
a positive colifonn result, the owner may be given another chance to
disinfect the well as it may take more than one attempt to effectively kill
off all the bacteria. If the owner refuses to treat the well or if the
problem persists, the owner will be asked to a pre-hearing conference to
discuss the problem and how it can be corrected. The Department of
Environmental Services has been consulted for ongoing problems with
water sources.
Conclusion
Inspection of the dairy plant did not reveal any definite routes for
contamination of the milk with Yersinia. No finished product or
environmental samples tested by the NHDPHS PHL were found to be
positive for Yersinia. A raw milk sample collected on 11/14/95 was found
to be positive for Yersinia enterocolitica, however, biochemically, this
isolate was not the same organism as the one implicated in the outbreak.
Although the final rinse of the bottle washer was not with treated water,
as required, there were no poor practices observed regarding bottle
washing and handling. The bottle washer exceeded all requirements for
time, temperature, and causticity. Environmmtal swabs taken from the
bottle washer were negative for Yersinia. A treatment system has been
installed on the water line for the final rinse of the bottle washer. No
water samples tested positive for Yersinia.
It was also determined that the employee responsible for the care and
feeding of the pigs was not engaged in the bottling of miik during
processing days. Employees are not permitted to go from the barns to the
processing plant, even on non-processing days, without changing their
clothing and footwear.
As previously stated, the NHDPHS BFP will increase surveillance of the
dairy plant. This oversite will include monthly inspections of the dairy
plant and monthly sampling of all finished product, raw milk, plant
operations water, and sweetwater.
Appendix A
Dairy Plant & Milking Facility*
Appendix B
Results of Finished Product Samples
collected by the Bureau of Food Protection
7/18/95 thru 11/13/95
Product Sampling NHPH
(SPC, Coli)
______________________________________________________________________________________________
Date Skim 2% Whole Choc 1/2 & 1/2 Cream Raw
7/13/95 <2500, <1 250, <1 <2500, <1 460, <1 4
(Quart) (1/2 Gal.) (Quart) (1/2 Liter) (LPC 260)
6/15/95 <250, <1 <250, <1 8,900
(Quart) (1/2 Liter) (LPC 1,100)
(SSC 700,000)
9/19/95 260, <1 540, <1 670, <1 350, <1 29,000, <1 400, <1
(Quart) (Quart) (Quart) (Quart) (1/2 Liter) (1/2 Liter)
10/24/95 1400, 1 2000, <1 2800, <1 15,000, <1 65,000
(Quart) (Quart) (Quart) (1/2 Liter) (LPC 24,000)
(SSC 360,000)
11/13/95 <250 <1 250 <1 18,000
(Pint) (Pint) (LPC 5,400)
(SSC 350,000)
October, 1995, Yersinia Outbreak Specimens
NH Public Health Laboratories
ENVIRONMENTAL SAMPLES
Collection Initial Final
Spec. # REF# Source (Type) Date Date Results
______________________________________________________________________________________________
MAC1 95EN509 Swab in enrichment media from 11/15/95 NEG NEG
front wall and floor area of
MAC delivery truck
MAC2 95EN510 Swab in enrichment media from 11/15/95 NEG NEG
side door and plastic louvred
curtains
MAC3 95EN511 Swab in enrichment media from 11/15/95 NEG NEG
rear door of MAC delivery truck
MAC4 95EN512 Swab in enrichment media from 11/15/95 NEG NEG
Front of bed - International
MAC5 95EN513 Swab in enrichment media from 11/15/95 NEG NEG
Side of International
MAC6 95EN514 Swab in enrichment media from 11/15/95 NEG NEG
Rear door of International
MAC7 95EN515 Swab in enrichment media from 11/15/95 NEG NEG
Can - creamery
MAC8 95EN516 Swab in enrichment media from 11/15/95 NEG NEG
Floor creamery cans
MAC9 95EN517 Swab in enrichment media from 11/15/95 NEG NEG
Crate
MAC10 95EN518 Swab in enrichment media from 11/15/95 NEG NEG
Floor crate
______________________________________________________________________________________________
Collection Initial Final
Spec. # REF# Source (Type) Date Date Results
______________________________________________________________________________________________
#1 95EN444 Swab in distilled water 11/10/95 NEG NEG
#2 95EN445 Swab in distilled water 11/10/95 NEG NEG
#3 95EN446 Swab in distilled water 11/10/95 NEG NEG
#4 95EN447 Swab in distilled water 11/10/95 NEG NEG
#5 95EN448 Swab in distilled water 11/10/95 NEG NEG
#6 95EN449 Swab in distilled water 11/10/95 NEG NEG
#7 95EN450 Swab in distilled water 11/10/95 NEG NEG
#8 95EN451 Swab in distilled water 11/10/95 NEG NEG
#9 95EN452 Swab in distilled water 11/10/95 NEG NEG
#10 95EN453 Swab in distilled water 11/10/95 NEG NEG
#11 95EN454 Swab in distilled water 11/10/95 NEG NEG
#12 95EN455 Swab in distilled water 11/10/95 NEG NEG
#13 95EN456 Swab in distilled water 11/10/95 NEG NEG
#14 95EN457 Swab in distilled water 11/10/95 NEG NEG
#15 95EN458 Swab in distilled water 11/10/95 NEG NEG
#16 95EN459 Swab in distilled water 11/10/95 NEG NEG
______________________________________________________________________________________________
HUMAN SAMPLES
Collection Initial Interim Interim Final
PHL REF# Source (Type) Date Date Results #1 Results #2 Results
_____________________________________________________________________________________________
95EN422 Stool, Cary Blair 11/06/95 NEG NEG NEG NEG
95EN423 Stool, Cary Blair 11/06/95 NEG NEG NEG NEG
95EN490 Stool, Cary Blair 11/14/95 NEG NEG NEG NEG
95EN491 Stool, Cary Blair 11/13/95 NEG NEG NEG NEG
95EN492 Stool, Cary Blair 11/12/95 NEG NEG NEG NEG
95EN493 Stool, Cary Blair 11/14/95 NEG NEG NEG NEG
95EN494 Stool, Cary Blair 11/13/95 NEG NEG NEG NEG
PHL REF# Source (Type) Collection Date Final Results
____________________________________________________________________________________________
95EN382 Isolate-Stool 11/03/95 Yersinia enterocolotica, biotype III
95EN411 Isolate-Stool 11/08/95 Yersinia enterocolotica, biotype III
Yersinia Outbreak Specimens
Public Health Laboratories/Environmental Micro Unit
October - November, 1995
Sample Type/ Collect. Phosphatase Coliform Y. ent. Y. ent.
Spec # Code Date Result Result Initial Final Result
(Water-MPN Result
/100ml)
______________________________________________________________________________________________
M1 Raw Milk 10-24 ------- ------- NF NF
M2 Well Water 10-24 ------- 5.1 ---- ----
M3 Sweetwater 10-24 ------- 1.1 ---- ----
M4 Skim Milk / 11-3 10-24 ------- 1 (not NF NF
confirmed)
M5 2% Milk / 11-3 10-24 Negative <1 NF NF
M6 Whole Milk / 11-3 10-24 Negative <1 NF NF
M7 Half & Half / 11-3 10-24 Negative <1 NF NF
X1 Skim Milk / 11-6 10-31 Negative <1 NF NF
X2 2% Milk / 11-6 10-31 Negative <1 NF NF
X3 Whole Milk / 11-6 10-31 Negative <1 NF NF
X4 Chocolate Milk/11-3 10-31 Negative <1 NF NF
X5 2% Milk / 11-6 11-01 Negative <1 NF NF
X6 Whole Milk / 11-3 11-01 Negative >120 Estimate NF NF
X7 Chocolate Milk/11-10 11-01 Negative <1 NF NF
X8 Water (1900ml) 11-02 ------- ------- NF NF
QM1 Well Water 11-14 ------- 3.6 ---- ----
QM2 Sweetwater 11-14 ------- <1.1 ---- ----
QM3 Raw Milk 11-14 ------- ------- NF + for Y.
Enterocolitica
QM4 Chocolate Milk/11-20 11-14 Negative <1 NF NF
QM5 Cream / 111-17 11-14 Negative <1 NF NF
MAC11 Sweetwater (2410ml) 11-15 ------- ------- NF NF
D1 Skim Milk / 11-17 11-16 Negative <1 NF NF
D2 2% Milk / 11-17 11-16 Negative <1 NF NF
D3 Whole Milk / 11-17 11-16 Negative <1 NF NF
D4 Chocolate Milk/11-17 11-16 Negative <1 NF NF
(NF = None Found)
(------- Denotes test not performed)
'CLINICAL' PIG SAMPLES
Collection Initial Interim Interim Final
Spec. # REF# Source (Type) Date Results Results #1 Results #2 Results
______________________________________________________________________________________________
Pig #1 95EN460 Tonsil, swab, in 11/10/95 NEG NEG NEG NEG
Cary-Blair
95EN461 Tongue, swab, in 11/10/95 NEG NEG NEG NEG
Cary-Blair
95EN462 Rectal, swab, in 11/10/95 NEG NEG NEG NEG
Cary-Blair
Pig #2 95EN463 Tonsil, swab, in 11/10/95 NEG NEG NEG NEG
Cary-Blair
95EN464 Tongue, swab, in 11/10/95 NEG NEG NEG NEG
Cary-Blair
95EN465 Rectal, swab, in 11/10/95 NEG NEG NEG NEG
Cary-Blair
Pig #3 95EN466 Tonsil, swab, in 11/10/95 NEG NEG NEG NEG
Cary-Blair
95EN467 Tongue, swab, in 11/10/95 NEG NEG NEG NEG
Cary-Blair
95EN468 Rectal, swab, in 11/10/95 NEG NEG NEG NEG
Cary-Blair
Pig #4 95EN469 Tonsil, swab, in 11/10/95 NEG NEG NEG NEG
Cary-Blair
95EN489 Tongue, swab, in 11/10/95 NEG NEG NEG NEG
Cary-Blair
95EN428 Rectal, swab, in 11/10/95 NEG NEG NEG NEG
Cary-Blair
Pig #5 95EN429 Tonsil, swab, in 11/10/95 NEG NEG NEG NEG
Cary-Blair
95EN430 Tongue, swab, in 11/10/95 NEG NEG NEG NEG
Cary-Blair
95EN431 Rectal, swab, in 11/10/95 NEG NEG NEG NEG
Cary-Blair
Pig #6 95EN432 Tonsil, swab, in 11/10/95 NEG NEG NEG NEG
Cary-Blair
95EN433 Tongue, swab, in 11/10/95 NEG NEG NEG NEG
Cary-Blair
95EN434 Rectal, swab, in 11/10/95 NEG NEG NEG NEG
Cary-Blair
Pig #7 95EN435 Tonsil, swab, in 11/10/95 NEG NEG NEG NEG
Cary-Blair
95EN436 Tongue, swab, in 11/10/95 NEG NEG NEG NEG
Cary-Blair
95EN437 Rectal, swab, in 11/10/95 NEG NEG NEG NEG
Cary-Blair
Pig #8 95EN438 Tonsil, swab, in 11/10/95 NEG NEG NEG NEG
Cary-Blair
95EN439 Tongue, swab, in 11/10/95 NEG NEG NEG NEG
Cary-Blair
95EN440 Rectal, swab, in 11/10/95 NEG NEG NEG NEG
Cary-Blair
Pig #9 95EN441 Tonsil, swab, in 11/10/95 NEG NEG NEG NEG
Cary-Blair
95EN442 Tongue, swab, in 11/10/95 NEG NEG NEG NEG
Cary-Blair
95EN443 Rectal, swab, in 11/10/95 NEG NEG NEG NEG
Cary-Blair
Go back to Misc. Milk References
Return to FDA Prime Connection Milk Safety references
Go BACK to the CFSAN food and consumer information pages
hypertext updated 7/30/96 by ear