Laboratory Cold Chain

7/17/2019 Laboratory Cold Chain

http://slidepdf.com/reader/full/laboratory-cold-chain 1/22

1

Laboratory cold chain quality performance –

an exploratory study

Joye DixonAuckland Distract Health Board, Auckland City Hospital, LabPlus Microbiology, Auckland New Zealand, [email protected]

Nihal P. Jayamaha, Nigel P. GriggSchool of Engineering & Advanced Technology, Massey University, Palmerston North, New Zealand,

[email protected], [email protected]

Abstract Assuring quality in a medical laboratory requires, among other things, acceptable levels of

quality of the incoming goods. The objective of this prospective study was to examine the performance of the delivery process of “cold chain laboratory products” through to the enduser and to identify opportunities for improvement. Temperature and delivery lead-timewere chosen as the desired quality characteristics. Pre-programmed temperature loggerswere used to log the data. The study was undertaken in New Zealand over the seasons ofwinter, spring and summer: June 2012 – February 2013. By using control charts and otherquality tools that facilitate "statistical thinking" it was found, in the District Health Board(DHB) studied, that the process of transporting cold chain products did not maintain thecold chain specifications. During the study it was discovered that products have a stability budget, which is the number of hours that a product can be outside its normal storagetemperature before it should be discarded because its quality is compromised and its shelf

life shortened. Recommendations have been suggested that may enable the institution toimprove its handling of the cold chain products that it purchases.

Key words: Laboratory Cold Chain, Statistical Thinking (ST), Control Charts, ProcessCapability Analysis

Topic: Quality and Performance

Methodology: Empirical work

Introduction Globalisation of world trade has made supply chains very efficient and effective. However,the resulting complexity of the supply chain can result in undesirable consequences on thesuppliers and the customers in the supply chain (Parker & Anderson.jr., 2002). It is notuncommon for New Zealand healthcare providers to procure pharmaceutical products andlaboratory products from as far away places as the USA and UK. The resulting supplychain involves many inland and offshore logistics services providers sharing informationthrough an integrated information system. While the quality of pharmaceutical andlaboratory products that can be supplied at ambient temperature levels (i.e. products whose

mailto:[email protected]













2

temperature specifications are within the ambient temperature levels that exist across thesupply chain) do not get affected by supply chain inefficiency and ineffectiveness, the samecannot be said about products whose cold chain has to be maintained (typically between2oC to 8oC).

There is a general consensus among practitioners that laboratory cold chain products donot receive the same treatment as the pharmaceutical cold chain products do, even thoughthe quality of the former (e.g. the quality of HIV AIDS test kits, reagents) can have assignificant an effect on the patients as the quality of the latter (Rodrigue, 2013; Walsh,2013).

This paper empirically examines the current state of the delivery process of laboratorycold chain products in terms of two quality characteristics — temperature and delivery leadtime — and provides short term recommendations to improve the quality performance ofthe supply chain, based on the said quality characteristics.

The literature supports the notion that the processes upstream of the local distributor(e.g. long haul flights) tend to preserve the cold chain and that only when the goods arerepacked and transported locally (via a specialised company) through to the final customer(laboratory) that breaches to the cold chain tend to occur (Department.of.Health.Victoria,2014; Rodrigue, 2013). Being an exploratory study, only the weaker link of the supplychain (the specialised delivery company inwards goods department of the district health board laboratory) was analysed by the lead researcher. Moreover, this is the link that thefinal customer – the laboratory – has the greatest control over to effect processimprovements in the short term.

Literature Review

Laboratory Cold Chains

As medicine developed through the centuries the advances created the basis for diagnosticdiscoveries. In the latter part of the 19th century the clinical laboratory started to beestablished. By the turn of the century many specific chemical, haematological and bacteriological tests had been developed due to innovations in basic science (Berger,1999a) enabling the clinical laboratory to be used for diagnostic purposes.

Different types of laboratories started to emerge, including physiological, pharmaceutical, forensic, public health, clinical chemistry and microbiological (Berger,1999b). Laboratories developed similarly in Britain and the United States of America(USA) due to the influences of medical faculties in Europe (Petts, 2012; Race, Tillery, &Dysert, 2004).

The number of laboratory tests available increased dramatically over the 20th century.The first half of the century was involved in developing individual biochemical methods forthe variety of parameters that medical professionals were interested in such as ionconcentration of elements in serum (Berger, 1999b). Because the refrigerator was notinvented until the late 1910s to early 1920s these methods did not require refrigeration.Today the range of different analyses available is now well over 3000, and still increasing,although not all will be available at each laboratory (Wians, 2009).

Requests for laboratory investigations increased as medical professionals found the datauseful in establishing diagnoses. The desire to have this data prior to treatment led to theexpectation of faster turnaround times for testing (Seligson, 1966), so more efficientmethodologies needed to be introduced. The first convenient and accurate kit-typediagnostic test, Clinitest®, was introduced in 1941 and measured reducing sugars in urine



3

(Chemical Heritage Foundation, 2010). This was followed by Clinistix® in 1956, the firstdipstick which measured glucose in urine (Chemical Heritage Foundation, 2010).Although Clinitest® has long been discontinued in modern laboratories dipsticks continuetoday with up to 10 parameters being measured by one dip in the urine. These types of tests

did not require refrigeration.Mechanisation followed with the first clinical autoanalyser being introduced in 1959(Berger, 1999b) and it was around this time that sub-ambient storage of some reagents became essential. Such analysers could perform multiple tests on much smaller samples ofthe patient’s blood (Seligson, 1966). In the 1960’s culture media became moresophisticated with antibiotics included as selective agents (Smith, 2005) necessitatingstorage at sub-ambient temperatures. The sixties also saw computers introduced intolaboratories in Europe, Britain and the USA but they did not become mainstream until the1970’s-1980’s (Keller, 1982). Miniaturisation of laboratory techniques began in the 1970swith microbiology and immunoassays being the initial beneficiaries’ and this has progressed through all disciplines (Fung, 2002; Hansen, Hardesty, & Myers, 1974; McGlennen, 2001). The reagents for these kit/reagents needed to be stored at sub-ambienttemperatures (Ammann, 2011; Ramanujam, Koelbl, & Ting, 1993). An example of thistype of kit is the API® range manufactured by bioMériuex Inc. More recently thesetechniques have moved onto full automation especially in larger laboratories (Wu, 2006).

The molecular era started in 1943 when Oswald Avery (1877-1955) demonstrated thatdeoxyribonucleic acid (DNA) played a significant role in the carriage of geneticinformation (Bersch, 2006). Research into DNA moved forward on multiple fronts withJames Watson (b. 1928), Francis Crick (1916-2004), Maurice Wilkins (1916-2004) andRosalind Franklin (1920-1958) all working on defining DNA’s structure in the early 1950’s(Bersch, 2006). In 1953 the peer reviewed academic journal Nature published this researchand the findings were widely accepted.

This along with further discoveries of critical enzymes and manipulation techniques ledto the development of new laboratory procedures. However, it was not until the inventionof the polymerase chain reaction (PCR) technique by Kary Mullis (b. 1944), in 1983, thatmolecular methodologies moved from the research laboratory into the clinical laboratory(Tang, Procop, & Persing, 1997) providing enhanced sensitivity and faster turnaroundtimes. Kits that use this technique require sub-ambient temperatures. Kits such as the HainLifescience Genotype® Mycobacterium CM/AS, Genotype® MDR TBplus and Genotype® SL require a mix of temperatures for storage. The primers and probes need to be kept at4oC and the master mix that contains the enzymes and dinucleotides at -20oC but are provided together in the same kit.Today, the diagnostic industry is starting to move towards "black box" type testing methods(Bersch, 2006) that includes the three main phases of PCR, extraction, amplification anddetection. This type of testing often does not require specialised staff or sub-ambientstorage conditions for reagents. An example of this is the Gene Expert® range of cartridgesmanufactured by Cepheid.

To complicate matters the supply chain's national and international regulations aremoving towards "controlled room temperature" (CRT) requiring additional monitoringsteps for ambient temperature packages, and adding complexity to the transportation process (Healthcare Commerce Media Corporation, April 09, 2010). Ultimately, even theambient temperature "black box" products being developed by the diagnostic industry willcome under the CRT auspices.



4

There is an abundance of literature and regulations on the development of drugs forhuman use including stability, experimental design and data analysis for the pharmaceuticalindustry but for the in vitro device (IVD) manufacturers, who manufacture kits for theclinical laboratory, there are guidance documents but few regulations.

However, the need for regulation was reinforced by a study performed in 2011 thatshowed that devices that had been deemed low risk and were not covered by regulationshad resulted in serious health problems or death for some patients (Zuckerman, Brown, & Nissen, 2011). A draft document was issued by the Food and Drug Administration (FDA)that covered companion diagnostic tests, with a deadline of 31 March 2013 but as late asOctober 2013 the FDA were still talking about regulating but with no definitive action todate (Novales-Li, 2013).

Currently, licenced laboratory developed tests (LDTs) and IVDs must be validated andany reagents used, including calibrators, controls, and sample diluents, must undergostability testing so consumers know the temperature and storage limits of the product. Inorder to deliver an accurate and precise result each of these components must befunctioning properly.

Stability studies cover the following: expected shelf life or expiration date, temperature,humidity and photostability, stress testing and storage conditions (Food and DrugAdministration, 2003; International Conference on Harmonisation, 2003). Environmentalfactors such as temperature, humidity and light can affect the quality and shelf life of a product. Stability testing using the climatic zones helps establish a shelf life and therecommended storage conditions for a product (Markens, 2009). The shelf life should be based on the long term study at the recommended storage condition. The product shouldalso be subjected to elevated temperatures for an appropriate time period to emulate theeffect of short term excursions outside the recommended storage conditions which occursduring shipping (World Health Organisation, 2009). The time-stress or excursion hours arecumulative and includes all stages of the product's life from production to storage in thecentral warehouse, then from the supplier to the distributor, which may mean exportingoverseas, customs or border processing, and lastly from the distributor to the customer. Atsigned acceptance of the delivery the stress period is ended for the distributor company as ithas no control over what happens post-delivery.

All products of a temperature-sensitive nature should be labelled appropriately to alertthe customer of the package’s storage requirements. The expectation is that the customerwould have procedures in place to manage the receipt of temperature-sensitive products.However, no products are sold with the cumulative stress hours available making it all themore essential that all parties involved in the transportation of goods monitor and ensurethat cold chain products are kept at the recommended temperature.

Knowledge of the stress hours originally available and how many have been used would be useful to the purchaser. For most products the process of bringing a cold chainkit/reagent to ambient temperature is a requirement before the test can be performed. Thespeed of chemical reactions are altered at different temperatures therefore the prerequisiteof bringing a kit to room temperature is necessary because the kit/reagent will have beendesigned to work in an ambient environment. These excursions should also be consideredas time-stress hours outside the recommended storage conditions. Consideration should begiven to how many times a kit/reagent is brought to, and left, at ambient temperature in thelaboratory environment and if consumption of the kit is slow then it should be discarded prior to finishing or reaching its expiry date.



5

Despite the numerous guidance documents surrounding the stability testing of pharmaceutical products and IVDs, they do not extend to stability in transit. Below (Table1) is a summary of the situation in 2011.

Table 1 – An Overview of Guidance Availability for Cold Chain Products (Seevers, 2011)

Cold Chain Products

Stability

Studies

ICH*

Q1A (R2)

(2003)

WHO

TRS 953

Annex 2

(2009)

Parental Drug

Association

Technical Report 39

(2007)

Pharma Industry

Survey (2009)

Long Term Guidance Guidance References ICH andWHO

Follow ICH

Accelerated Guidance Guidance References ICH andWHO

Follow ICH

Temperature

Excursion

No

Guidance

No

Guidance

-20oC for 2 days

40o

C/75% RH for 2days

-70oC for 36 months

-15o

C to -25o

C for 2weeks to shelf life0-2oC for 2 weeks to24 months30oC/65% RH or30oC/75% RH for 2weeks to 6 months40oC/75% RH forless than 2 weeks to 6months

Temperature

Cycling

No

Guidance

No

Guidance

-20oC for 2 days

followed by25oC/60% RH for 2days(Repeat for a total of3 cycles)

Temperature cycling

and/or freeze/thawfor less than 2 weeksto 1 month.

* ICH: International Conference on Harmonisation WHO TRS: World Health Organisation Technical Report Series

While the product is being produced and then housed in the manufacturer's warehousecontrolling the cold chain is simple because the manufacturer will have highly controllableinternal systems in place. However, the aim is sell and distribute products and this activity

is difficult to control because it may involve several different organisations.At the DHB studied pharmaceutical products such as drugs and vaccines which areintroduced into patients are transported in portable insulated containers (PICs) fromsuppliers to the IGs department. In packages carrying a critical temperature-sensitive product, such as vaccines which lose potency outside the recommended temperature range, a data logger is included. Laboratory kits/reagents and culture media are "packed cold" 1

1 Products are removed from the 2-8oC cold room and packaged together and with paper to help retain the required

temperature.



6

and transported in boxes with appropriate labelling such as "Refrigerate" or "2oC to 8oC".The box may or may not be labelled and/or insulated.

There is no time constraint on freighting non-urgent cold chain packages within theindustry. A courier company collecting a package from a supplier in the morning does not

guarantee that it will be delivered the same day. Packages picked up late in the day willalmost certainly be left at the depot overnight or possibly even in the courier van. Welllabelled packages may make it into a cold room at the depot but many will not. Somedepots may not have a cold room.

The cold chain industry has some concerns over "the last mile" or the end stage in thelogistics chain, where a shipment meets its ultimate destination. This may be a laboratory, a pharmacy or it may be the patient. All temperature-sensitive products are at risk if handledimproperly.

The Inward Goods (IGs) department is the first step in the last mile. Before goods areaccepted for storage, quality control requirements need to be met. This involves ensuringthe product is in an undamaged condition and that temperature-sensitive products areactually at the correct temperature (Learning and Skills Improvement (LSIS), 2013). If a product has a 2oC-8oC temperature range then it should be close to or within that range ondelivery. In some countries probes can be used to check between products on a pallet. Ifthe truck is refrigerated then the vehicle temperature gauge reading should be recorded. The products should have temperature-sensitive labels on the outside of the box.

On receipt the products should be divided into ambient, refrigerated or frozen. Therefrigerated and frozen products should be placed in a cold room as a matter of priorityuntil the paperwork has been processed.

Once processed the IGs department should have a standard operating procedure (SOP)for delivery, which stipulates maintenance of the cold chain, to the relevant departments.Records should be kept of the time that products are delivered to the various departmentsand kept on file so that the last mile is monitored to the very end.

Prior Research on Pharmaceutical Cold Chains

The literature review indicated that most of research on cold chain risk management in themedical industry has been on biopharmaceutical products. These studies included researchon quality risks associated with accidental freezing of vaccines (e.g. (Matthias, Robertson,Garrison, Newland, & Nelson, 2007; Nelson et al., 2007; Nelson et al., 2004), quality riskawareness assessment among medical staff (e.g. (Thakker & Woods, 1992), real timetemperature monitoring using state-of-the-art technology to predict cold chain performance(e.g. (Chen & Shaw, 2011), review of policy documentation and cold chain management best practices for transportation (e.g. (Bishara, 2006; Soeung et al., 2004)) and so on. Someof these studies are discussed in turn.

Matthias et al (2007) conducted a systematic review on extant literature on riskassessment of accidental freezing of vaccine to identify the prevalence of vaccine freezingin the cold chain. They found that among studies that examined refrigerated transports, between 14% to 35% (the figure varied from study to study) of the shipments were exposedto accidental freezing, while among the studies that examined all segments of the supplychain—depending on the study that they reviewed—between 75% of 100% vaccineshipments were exposed to accidental freezing (among the studies that they examined, theyfound that the more rigorous studies reported higher figures). Nelson et al (2007)monitored cold chain temperatures on a randomly selected sample of vaccine shipments



7

across Bolivia to identify which segment of the supply chain exposes the greatest risk to thevaccine cold chain. They found that the greatest quality risk (exceeding the 2oC to 8oCspecification) occurs in the last leg of the shipment of goods to the customer.

Thakker and Woods (1992) studied the extent to which United Kingdom (UK) medical

staff adhere to departmental guidelines on vaccine cold chain maintenance, based on asurvey of 40 respondents representing 40 general practices (only the districts Manchesterand Bradford were covered). They found that only 16 out of 40 respondents were aware ofthe correct guidelines on cold chain maintenance. Similar findings were reported by Nelson et al (2004) on a cold chain elsewhere (Indonesia), although accidental freezing wasthe main issue that they investigated. The researchers also encouraged other researchers toconduct stability studies to determine whether maintaining a cold chain at an elevatedtemperature (this reduces accidental freezing, energy costs, and waste) can be justified froma health perspective.

Chen and Shaw (2011) showed how state-of-the art automated data collection methods(e.g. radio frequency identification techniques) can be used to feed in cold chain producttemperatures to a central station which can take swift action based on temperature trends.The quality tools that they used to monitor data were the exponentially weighted movingaverage (EWMA) control chart and artificial neural networks.

Prior Research on Laboratory Cold Chains

In regards to temperature sensitive medical laboratory goods transportation anddistribution, it appears that current research is focused more towards robust product designstrategies (i.e. making products insensitive to temperature variation) (e.g. (Foo et al., 2012; Hoizey et al., 2005) and “point-of-care technologies” (Lee et al., 2010; Shott, Galiwango,& Reynolds, 2012).

Robust design refers to designing a product at a lower cost—for example, by changingthe product’s design parameter settings rather than spending money on expensive designcomponents (e.g. raw material) or excessive maintenance costs (e.g. cold chainmaintenance over long haul overland transport)—so that the product’s performance (in alaboratory product context, the product’s efficacy) remains robust against the variations ofthe environmental factors such as the temperature and the humidity (Montgomery, 2013; Roy, 2010). Foo et al. (2012) as well as Hoizey (2005) argue that maintaining the coldchain for long haul shipments in developing countries (e.g. across sub-Saharan Africa) iscostly and robust designs remain an attractive proposition. A similar argument is given for bringing the laboratory to the patient (the term point-of-care technologies refer to thecollection of technologies that enable medical care, including laboratory services be provided near to the patient).

New Zealand is a developed country and maintenance of the cold chain of laboratory products is not cost prohibitive. Whilst alternative technologies such as robust designs and point of care technologies should be explored, in the short term, efforts should be directedto maintenance of the cold chain.

Statistical Thinking

There are numerous statistical methods involving “number crunching, distributions andcontrol charting” (Britz, 2000) used by statisticians on a daily basis but these are moreeffective if used in conjunction with statistical thinking (ST). ST is about looking at timeseries data to elucidate the process or phenomenon being observed: what has happened in



8

the past, what is happening currently and what should happen in the future (Ramsey &Schafer, 1997; Wild & Pfannkuch, 1999).

W. Edwards Deming popularised ST among Japanese workers and managers from 1950sthrough to 1970s when he was helping the Japanese industry to improve quality and

productivity to become a competitive industrial nation post World War II (JUCE, 2014). ToDeming, ST is a way of understanding what is happening to a process. He showed theJapanese engineers and managers that on many occasions processes do show natural(random) variation (he called this common cause variation), which do not warrant action.He called such processes stable and predictable processes. Deming showed that a workerignorant of this important statistical property would make an unnecessary adjustment to a process (thinking that an adjustment is needed to curb the variation) when leaving the stableand predictable process alone would have been the right option. Deming showed thatunnecessary adjustments — which he called “tampering” — worsens process variation,causing more loss to the organisation (more rework and waste) (Deming, 1982; Rao et al.,1996). Deming identified two instances where action is necessary: (a) if the variation doesnot look random (he showed that this occurs due a cause that is not normally part and parcel of the system, which he called a “special cause”), and (b) if the natural variation ofthe stable process (common cause variation) is so excessive that the process becomesincapable of meeting the customer specifications (Deming, 1982; Rao et al., 1996). ST iswidely regarded as one factor that made Japanese products superior to (high quality at acompetitive price) the products manufactured in the west in the 1980s (Hoerl & Snee, 2012; Snee, 1990). More recently an official definition was given to ST.

ST was defined by the American Society for Quality as “a philosophy of learningand action based on the following fundamental principles:

1) All work occurs in a system of interconnected processes.2) Variation exists in all processes.3) Understanding and reducing variation are keys to success.” (American Society

for Quality, 1996).The research reported in this paper uses ST to make recommendations to the

process owners to improve the cold chain delivery processes.

Research Questions

Given the background provided on cold chain management and the benefits of ST, thefollowing research questions are raised:

RQ1: Given the portion of the supply chain studied and the quality characteristics chosen,does the delivery process remain stable and predictable?

RQ2: Given the portion of the supply chain studied and the quality characteristics chosen,is the delivery process capable of meeting the requirements placed upon it?

RQ3: If the capability of the process needs to improved, what short term interventionscould be put in place to improve process capability?



9

Methodology

Demarcation of the System Boundaries and the Critical Control Points

As mentioned earlier, only the local movement of the cold chain goods — that is, within New Zealand transport from the New Zealand distributor through to the hospital medical

laboratory, the weaker link as far as preservation of the cold chain is concerned — wascovered. Figure 1 shows the particular system studied (system 2).When cold chain goods move through the system studied (System2 in Figure 1), the

temperature of the goods enclosed within a package rises due to exchange of heat with theambient. As such, the temperature of the product does not remain stable (there is an upwardtrend in the temperature particularly in summer months) throughout the delivery process.

Figure 1: Demarcation of the system studied within the overall delivery system

However, given that the temperature of the goods entering the laboratory is expected to be within the temperature range specified for cold chain products (2oC – 8oC), it is possibleto assess the stability (predictability) and the capability of the system based on thetemperature of the goods at the point of arrival to the laboratory. Thus this point was takenas a critical control point for data monitoring. Figure 2 depicts the critical control pointsidentified for the system studied, while Table 1 provides a description of these critical points.

Figure 2: The subsystems of the system studied and the critical control points

Table 2: Definitions of the Critical Control Points

CriticalControl Point

Description of the Critical Control Point

A Goods are packaged and wait for Courier pickup to leave the supplier



10

B Goods enter Inward Goods department

C Goods transported from Inward Goods department to laboratory

D Goods delivered around laboratory via trolley

E Goods unpacked in the laboratory

Note: (a) Please interpret this table in conjunction with Figure 2.(b) Temperature and elapsed time monitoring was sporadic at critical points C

Given that the highest temperature occurs at control point E, a control chart wasconstructed for the temperature at point E, based on the temperature of the incoming goodsin each trip. The I-MR (Individuals-Moving Range) chart was considered to be mostappropriate chart as the data of the variable were individual data (Wheeler & Chambers,2010).

In addition to the I-MR chart for the temperature at point E, the stability/predictability ofthe delivery process was also assessed in terms of the delivery lead time (from point A to point E) using a run charts (each data point in run chart representing one trip). The controlcharts, the run charts as well as process capability analysis was accomplished using theMinitab® 16 software package.

Selection of Participating Delivery Companies and Temperature Data Logging Equipment

A mix of companies were chosen to take part in the survey. They ranged from a familyfirm, to companies that specialised in medical laboratory point-of-care kits to largemultinational companies. The large multinational companies included a pharmaceuticalcompany and two distributor companies involved in selling a wide range of products andmachinery. All the companies made regular deliveries to the laboratory.

Either the Warehouse Manager or the Operations Manager were contacted within eachcompany to explain that the research being undertaken was focusing on the 'last mile'.Written instructions were left with each company. It was agreed by all parties that thecompanies would remain anonymous.

Two companies were approached for the supply of loggers. Temprecord ™ InternationalLtd is an IANZ accredited company whose core competency is the production of a range oftemperature and humidity recording solutions for industry including software, calibrationand support. Six "scientific multi-use" Temprecord ™ loggers were received, each of whichwas already calibrated and with the correction factor built into the logger along with atraceable calibration certificate. The loggers have a long life battery (two year guarantee), a32K memory, are calibrated to within ± 0.2oC, readable to two decimal places and are programmable between two seconds and 36 hours in two second increments. Each loggerhas a start delay, start, stop and marker buttons, and inside and outside range indicators butcurrently no LED display (Temprecord™ International Ltd, 2010).

Bell Technology Ltd is regarded as a leader in the supply of industrial process,laboratory and portable instrumentation in New Zealand. Six "ESCORT Intelligent MINI(Multi Trip)" loggers were received all of which were not calibrated. The loggers have along life battery (one to two year guarantee), memory sufficient for 1868 samples, internalsensor accuracy ± 0.5oC, programmable between one and 255 minutes in one minuteintervals and readable to one decimal place. Each logger has a LED display with start and



11

stop buttons and time spent under or over specification (Escort Data Loggers Inc., 2010).The ESCORT loggers were calibrated and found to have a correction factor of -0.5oC. Thiscorrection could not be built into the logger therefore it had to be subtracted from eachreading.

Both companies have user-friendly software to download the loggers’ information.Graphs and data were able to be produced for each trip. Temprecord ™ data could beexported to Excel for manipulation.

Results Interpreting the I-MR Chart and Process Capability Analysis

Figure 3 depicts the control chart ( I-MR) and process capability information. None of theindividual observations on the moving range ( MR) chart fall outside the control chartlimits. There were several trips close to the upper and lower limits and a run of sixconsecutive runs above the average but they were not rising or falling continuously.Otherwise, the points displayed a random pattern. Since the variation ( MR) chart was incontrol the individual control chart ( I chart) was examined. The Individual chart showedtwo points outside the upper and lower control limits implying possible a special causevariation that warrants further investigation.

Figure 3: The control chart and process capability information for the trips studied before

removing the unusual observations

The first trip (trip # 12) that implied instability (out of control) was from thecompany that encased the logger in an ice slurry which is not the usual method of



12

transporting cold chain products! Consequently, this was treated as an assignable cause. Itis probable that because the shipment was being monitored with a logger the companydecided to ensure it remained between 2oC-8oC. Unfortunately, the temperaturereached well below 2oC, and if this product had been a vaccine, it would have been

rendered inactive!The second trip that implied instability was found to be the result of product beingstored overnight at the courier company's depot at ambient temperature. The trip took place between 30-31 January 2013 when the ambient temperature averaged a high of26 oC and a low of 16.5 oC (AccuWeather Inc., 2013). The trip from "supplier toInward Goods" took 18 hours.

The graphs were reproduced with the datum from the first unusual observation (trip #12) removed as this was a special cause and not a normal practice but the datum from thesecond trip was retained as this happens regularly and is an area that could be improvedupon. The I-MR chart and process capability information (Figure 4) reproduced byMinitab® showed little change in the indices. Although the estimated process meanincreased from 9.98 oC to 10.28 oC while the spread of data (standard deviation) wasreduced from 3.298 to 2.869 (2.9 oC to the 1dp). The normal probability plot in Figure 4implies that the temperature data does not follow a strict normal distribution.

Figure 4: The control chart and process capability information for the 36 trips studied after

removing the unusual observation

The estimated process mean (after removing trip # 12) was found to be 10.28oCwhile the estimated process standard deviation was found to be 2.87oC (the figure



13

shown in the left had side box of the capability plot shown in Figure 4) . This standarddeviation was treated as the baseline figure for future process improvement (reducing thevariation and moving the target value towards 5oC).

Figure 4 shows that the capability indices Cp and Cpk values for the process studied

were 0.26 and – 0.23 respectively. Cp and Cpk for a just capable process is treated as 1.00 (anyvalue below this shows incapability in meeting the specifications), whilst Cp and Cpk values (ingeneral Cpk < Cp) for a good process is (the industry standard) is considered to be > 1.33(Montgomery, 2009; Rao, 1996). These figures indicate that the process under review wasnot capable of meeting the expectations / requirements placed upon them by thecustomers: laboratory scientists and technicians. One way to improve the capability of the process is to improve the quality of packaging (e.g. provide better insulation, include arefrigerant). When the quality of packaging is improved, the average temperature of thegoods arriving the laboratory does reduce, because less heat is exchanged with theambient. It is also possible to reduce the average temperature of the goods arriving thelaboratory if the average delivery lead time can be reduced.

Interpreting the Run Charts for Delivery Lead TimeFor the purpose of data analysis, the delivery lead time, the time elapsed from criticalcontrol points A to E (Figure 2), was partitioned into two elapsed times: elapsed time forgoods movement from point A to B (supplier to inwards goods department) and elapsedtime for goods movement from point B to E (inwards goods to the laboratory).

Figure 5: Minitab® run chart for time elapsed "supplier to inward goods department”

Both Figures 5 and 6 show that whilst there are no clustering, mixtures, trends andoscillation of data (see Wheeler and Chambers, 2010 or Evans, 2013 for run chart rules)there are a number of so called “astronomically high” elapsed times for cold chain goodsmovement, which suggests that the delivery lead time is unstable and unpredictable.Comparing Figure 6 with Figure 5, it becomes evident that most of the spikes of high



14

elapsed times occur when the goods move from the inwards goods department to thelaboratory. The normal probability plot shown in Figure 7 shows that there are about nineastronomically high elapsed times out of 36 (25% of instances) trips (elapsed times)studied. The root causes for the aforesaid spikes were investigated along the five Ms: Menand Women, Money, Methods, Material, and Mileau (mother nature, surroundings, settingetc.) (Rao et al., 1996; Wilson, Dell, & Anderson, 1993).

Figure 6: Minitab® run chart for time elapsed "inward goods department to the laboratory”

25002000150010005000

99

95

80

50

20

5

1

Elapsed Time (min)

P e r c e n t

Probability Plot of Elapsed Time (min) Assumed Distribution : Normal

Figure 7: The normal probability plot to detect astronomically high elapsed times for delivery of

cold chain goods from the "inward goods department to the laboratory"

Astronomically high values!

(9 of them)



15

Figure 8 shows the cause and effects diagram drawn for the high delivery lead time andtemperature variation (since the two effects ‘delivery lead time’ and ‘temperature’ arecorrelated, it was decided by the researchers that a single cause and effect diagram wouldsuffice).

Figure 8: The cause and effects diagram

The delivery goods from the inwards goods department to the laboratory are handled bythe inwards goods department. Therefore improvement of the work processes of theinwards goods department stands out as a high leverage opportunity for processimprovement in order to reduce the delivery lead time (and the associated variation). When

Measurement Materials Method

Mileau (Environment) Men and Women Money (Capital)

Excessive

Delivery Lead

Time and High

Temperature

Variation

Problem

Statement

Insufficient staff in Inward

Goods department to turn

product around rapidly

Inadequate training of Inward

Goods staff in Good

Distribution Practice and

awareness of the Cold Chain

Cost of portable insulated

containers relative to the cost

of the product

Laboratory delivery staff have

many tasks and may not be

available when product

delivered by Inward Goods

Supplier staff do not

accommodate cold chain

products once packaged for

delivery

Supplier staff rely on 'stability

budget' for out of specification

deliveries.

Laboratory products are left in

a North facing fo yer until

delivery to co ld rooms within

laboratory

Packages left at room

temperature whilst being

delivered - delivery may take

a long time

Inward Goods staff not aware

of the 'Cold Chain'

Inward Goods staff accept

cold c hain products regardless

of condition

Courier drivers do not adhere

to cold chain requirements

Inward Goods do not

accommodate varyingtemperature requirements of

products

Products not packaged

adequately for delays in

delivery both by Courier and

from Inward Goods to

Laboratory leaves products at

room temperature all day 'just

in case' they need to be used

Products packed and left in

Supplier warehouses until

pickup can reach high

temperatures in summer

Warehouse type environments

such as Inward Goods

departments c an reach high

temperatures in summer

Courier vans do not have

refrigerated compartments

Inward Goods department

hasn't been refurbished for 38

years.

Inward Goods staff not trained

in Good Distribution Practic e

Cold Chain products may be

'packed cold' which does notaccommodate delays in

delivery

Cold Chain products may not

be packaged in insulated

containers

Products vary in temperature

requirements

Industry is bound by guidance

documents for laboratory

products rather than

regulations

Packaging often has confusing

labels or no labels related to

Cold Chain

Supplier packs products which

then wait for c ourier pickup at

room temperature

Product not delivered directly

to Inward Goods by Courier

van - may take many hours to

deliver

Product may spend overnight

at room temperature in Courier

depot

Inward Goods cold room doesnot have its temperature

monitored

No specifications around lead

time for delivery of uninsulated

Cold Chain products

Data logging not performed on

laboratory cold chain products



16

the lead time is reduced and made stable, the average temperature (and its variation) ofgoods arriving at the laboratory will also reduce and this will therefore improve thecapability of the goods movement process overall.

Discussion and RecommendationsIf "Quality health care means doing the right thing, at the right time, in the right way, forthe right person – and having the best possible results" (Agency for Healthcare Researchand Quality, 2007) then it is one of a DHB’s core responsibilities to undertake riskmanagement for cold chain products, their processing through IGs and delivery to variousdepartments.

This study covered the delivery of product from suppliers to a large tertiary institutionlaboratory via the IGs department. As can be seen from the results the cold chain is notupheld during most trips. Cold chain products for the laboratory are not treatedsimilarly to pharmaceuticals by distributing companies. The cold chain around pharmaceuticals is more stringent than for laboratory products and as a result drugs andvaccines are transported in portable insulated containers (PICs) with the appropriaterefrigerant whereas laboratory products are packaged in boxes with 2oC – 8oC labelsaffixed. A refrigerant may or may not be present, the box may be or may not be insulated.

The IGs department did not record the delivery time of packages and as a result therewas no proof of delivery of a package. Neither did it separate cold chain and ambient products to be independently processed, stored or delivered. Consequently, product sat atambient temperature for various time frames thereby adding to a product’s cumulativeexcursion hours and shortening a product’s usable shelf life within the laboratory.

The question is – why are laboratory cold chain products not treated similarly to pharmaceuticals by suppliers? Although products are not being introduced into a patient,the results of diagnostic testing are being used in up to 70% of diagnoses (Hallworth, 2011)and therefore ensuring that the right result (i.e. lowest number of false positives and falsenegatives) is obtained is essential. If the products have not been stored correctly duringtransit, especially once they arrive at the IGs department, then the quality of the productcould be compromised.

Based on the lead author’s experience and her own investigations on the potential rootcauses for the reduced priority for laboratory cold chain products compared to pharmaceutical cold chain products (in terms of maintaining the cold chain), the following potential root causes are proposed:

• Money - High Cost of Portable Insulated Containers (PICs) Relative to the Costof the Product Supplied: Reusable PICs could be used and a recycling programme developed. When a new delivery is made the PICs from the previous delivery to IGs could be picked up and returned to the company ontheir next pick up. This would be made easier by companies having arelationship with a courier or logistics company as suggested as GoodDistribution Practice (GDP). There could even be a city-or-countrywidecycling of PICs where companies that receive PICs use them for the next trip.

• Procedures - The industry is bound by guidance documents as opposed to

regulation. However, this is changing as governing bodies realise that



17

maintaining the cold chain is important and the logistics industry gathersmomentum to improve their service in this area.

• Communication – It appears that the IGs department has not necessarily kept up

with changes in the logistics area. At the very least such a department shouldrecognise the difference in the handling requirements of cold chain to ambienttemperature products and have access to a well maintained, monitored coldroom. This may mean a change to current practices. However, as well asmanaging the cold chain, the future predicts such areas will have to be climatecontrolled due to "controlled room temperature" guidance documents presentlyin the pipeline.

The following short term process improvement recommendations were proposed tothe management of the DHB.

Recommendation #1

Health Alliance (hA) should notify their suppliers that packages should be labelled as tothe appropriate storage temperature on the outside of the box. Inappropriate labelsshould be occluded on reused packaging and mixed packages should be discontinued.IGs should not accept goods that have confusing storage condition labelling.

Recommendation # 2

Health Alliance should insist that all cold chain packages should be delivered in arefrigerated vehicle or container. Such pressure would ensure that distributioncompanies complied and the courier market would respond in kind. Also, the arrival of a product in such a container would reinforce to the IGs staff that the product requires specialattention.

Recommendation # 3

The IGs department should record the delivery time of packages. This is common practice in warehousing and is important for GDP. It also provides an audit trail fordeliveries. It may be as simple as writing on the box, in felt tip, the delivery date and time.This information could then be entered into the supply chain computer software when the product is receipted in.

Recommendation # 4

Packages need to be sorted on arrival onto ambient temperature and cold chain trolleys.Cold chain trolleys would then be placed in the cold room until ready to be processed.

The cold chain trolleys need to be smaller so that only a limited number of packages areremoved for processing at any one time. Once processed the cold chain trolley would bereturned to the cold room until delivery to the laboratory was imminent. To make this alean process may require some reconfiguration of IGs so that there is a better work flow between the staff and the cold room.

Recommendation # 5

Colour coding the packages as to the department they are to be delivered to within thelaboratory would assist the delivery staff, ensuring that they always deliver the products



18

to the correct department. This would help prevent "lost packages" that may not behandled correctly by the receiving department if it is a product they do not normallyreceive. It would also help when there is more than one box per picking slip.

Recommendation # 6A cold chain trolley for "Frozen" products should also be available and separate to the products to be refrigerated. Such products are usually packaged in dry ice which isregarded as a dangerous good. As a result the package will be labelled with the appropriatecategory labels making it obvious it is a cold chain product. Dry ice will keep the productfrozen but it would be useful to separate such products so that it is not left in the coldroom over a weekend as the dry ice will dissipate leaving the product at 4 oC rather than -20oC. "FROZEN" stickers should be used on these products.

The "REFRIGERATE" or "FROZEN" stickers on the picking slip are useful to thestaff unpacking the products and should remain but it may also be useful to put anothersticker, in plain sight, on the package especially if the package does not have any other

indication that it is a cold chain product.

Recommendation # 7

The cold room should have temperature spatials 2

done six monthly to ensure that it isrunning at the correct temperature range.

Recommendation # 8

Staff should be trained in all aspects of warehousing so that they can troubleshoot whenthere are problems. This training would also give them an insight into the logistics of thecold chain and their role in ensuring it is not broken. It would also give them the ability torecommend changes that would result in quality improvement of the process.

Recommendation # 9

The delivery venue at the laboratory is unacceptable for cold chain products. Not only are products left in a public space but it is a north facing glass fronted foyer which becomesvery hot in summer. The foyer area should have a small cold room installed, perhaps underthe stairs, where the cold chain trolleys could be stored until delivery staff are available todistribute the products to the departments. Whilst in this cold room the packages could besorted into departments using the department colour coding placed on the package by IGs.Then the products could be delivered directly to each department and placed in the differentcold rooms.

Recommendation # 10

Cumulative excursions need to be managed. An example of a possible method follows(Table 2).

2 Spatials – data loggers are placed at various points around the cold room and the temperature logged over a specific time

frame to give an indication as to any hot or cold spots in the cold room and whether the temperature remains within thecorrect range.



19

Table 2: Example of Check Sheet for Temperature Excursions on Product

Date and Time Hours atAmbient

Temperatur

Accumulated Hours

Removed from ColdRoom

Returned to ColdRoom

Grouping Kit

Batch #: C1234Received:17/1/13 Opened:15/2/13

15/2/13 0800 15/2/13 1500 7 716/2/13 0800 16/2/13 1300 5 1217/2/13 0800 17/2/13 1600 8 20

The practice of removing a product from the cold room in the morning and leaving it atambient temperature all day may need to be adjusted. Labels could be affixed to each product which can be used to monitor the excursions. If the requisite excursion hours arenot available to the laboratory an average arbitrary figure could be used such as 120 hours.

This label could then be photocopied or scanned and stored as an audit trail for that particular product.

Recommendation # 11

The laboratory’s mission statement and values should be displayed in each departmentand staff reminded of them regularly. All staff, no matter what their role, should betreated with respect by other staff and only then will the laboratory’s and the DHB’svalues be upheld.

Staff need to be reminded by senior management that as personnel of the laboratorythey are all on the same team and that no one person is superior to another. For thelaboratory to fulfil its mission statement people should be treated with respect regardless

of their role. Staff need to remember that the delivery staff are essential as they supply products to the laboratory staff to ensure testing is done in a timely manner.

Conclusion

Customers of the laboratory services such as clinicians rely on true and accurate results.Companies supply laboratories with products to perform tests and investigations using bothcold chain and ambient products. Cold chain product packages need to meet thechallenging temperature specification of 2oC to 8oC, the target being 5oC. Breaches of theupper specification limit over prolonged periods can have an adverse impact on the qualityof laboratory services through inaccurate test results.

This study was undertaken to determine whether the portion of the supply chain studied,

and using the quality characteristics chosen, was able to provide a stable and predictabledelivery system which met the requirements of the products being transported. If capabilityneeded to be improved what short term interventions could be put in place?

The study found that the ambient product process was stable, predictable and met thetransport requirements the same could not be said about the delivery of cold chain products.Based on the I-MR charts, the cold chain delivery process was almost always found to bestable but the process means were way off-target and the variation was too high, given thetemperature specification.



20

The cold chain problem arises because distributing companies transport pharmaceutical products differently to laboratory kits and reagents (IVDs). This is due to the regulationsurrounding pharmaceuticals compared to the guidance documents for products used inlaboratories. Despite the fact that the kits and reagents are not used to treat patients they

are used to help diagnose patients’ conditions and treatment so it is essential that theirquality is not compromised. Data loggers are the most efficient way of loggingtemperatures as they are reusable and their data can be downloaded to a computer or thecloud, stored indefinitely and provide an audit trail.

According to the literature the last mile is of concern to the transportation anddistribution companies as they no longer have control of the product and it is at this pointthat the cold chain often breaks down.

It is hoped that the results will stimulate a larger and more comprehensive survey withall of the critical control points being monitored through co-operation between distributors,courier companies and the DHB. With a trip defined by these critical control points,moving ranges could be calculated for each part of the trip creating the opportunity togenerate key performance indicators (KPIs) and it would also define the times when the product was sitting in a warehouse as opposed to being transported. This would giveincreased information by clearly showing which areas of the transportation process requiremodification thereby strengthening the need for change in various areas of the supply chaineither within the hospital, courier depots or by the supplier.

The study showed how statistical process control can help in identifying delivery performance gaps of cold chain products to a laboratory and uncovered the need to identifyindustry best practices related to delivery of laboratory cold chain products.

References

American Society for Quality. (1996). Glossary and Tables for Statistical Quality Control: Quality Press.Ammann, C. (2011). Stability Studies Needed to Define the Handling and Transport Conditions of Sensitive

Pharmaceutical or Biotechnological Products. AAPS PharmSciTech, 12(4), 1264-1275. doi:10.1208/s12249-011-9684-0

Berger, D. (1999a). A brief history of medical diagnosis and the birth of the clinical laboratory: Part 1--Ancient times through the 19th century. Medical Laboratory Observer, 31(7), 28-30, 32, 34-40.

Berger, D. (1999b). A brief history of medical diagnosis and the birth of the clinical laboratory: Part 2--laboratory science and professional certification in the 20th century. Medical Laboratory Observer,

31(8), 32-34, 36, 38.Bersch, C. (2006). A Brief History of Medical Diagnosis and the Birth of the Clinical Laboratory: Part 5a -

The foundation of molecular science and genetics. Medical Laboratory Observer .Bishara, R. H. (2006). Cold chain management - An essential component of the global pharmaceutical supply

chain. American Pharmaceutical Review, January-February, 1-4.Britz, G. C. (2000). Improving Performance Through Statistical Thinking: ASQ Quality Press.Chemical Heritage Foundation. (2010). Helen M. Free and Alfred Free. Chemistry in History. Retrieved 8

June 2013, from http://www.chemheritage.org/discover/online-resources/chemistry-in-history/themes/pharmaceuticals/diagnosing-diseases/free-and-free.aspx

Chen, K.-Y., & Shaw, Y.-C. (2011). Applying back propagation network to cold chain temperaturemonitoring. Advanced Engineering Informatics, 25(1), 11–22.

Deming, W. E. (1982). Quality, productivity and competitive position. Cambridge, MA MassachusettsInstitute of Technology, Center for Advanced Engineering Study.

Department.of.Health.Victoria. (2014). Cold chain management. Retrieved 01 March, fromhttp://www.health.vic.gov.au/immunisation/cold-chain-management.htm

Escort Data Loggers Inc. (2010). Escort Data Loggers Inc. Simple and Smart. Retrieved 12/6/13, fromhttp://www.escortdataloggers.com/brochures/iMINI/Escort_Intelligent_MINI_Multi_Trip.pdf

http://www.chemheritage.org/discover/online-resources/chemistry-in-history/themes/pharmaceuticals/diagnosing-diseases/free-and-free.aspx




http://www.health.vic.gov.au/immunisation/cold-chain-management.htm


http://www.escortdataloggers.com/brochures/iMINI/Escort_Intelligent_MINI_Multi_Trip.pdf








21

Foo, P. C., Chan, Y. Y., Too, W. C. S., Tan, Z. N., Wong, W. K., Lalitha, P., & Lim, B. H. (2012).Development of a thermostabilized, one-step, nested, tetraplex PCR assay for simultaneousidentification and differentiation of Entamoeba species, Entamoeba histolytica and Entamoeba disparfrom stool samples Journal of Medical Microbiology September 61(9), 1219-1225

Food and Drug Administration. (2003). Guidance for Industry: Q1A (R2) Stability Testing of New Drug

Substances and Products. Retrieved 13 June 2013, fromhttp://www.fda.gov/downloads/regulatoryinformation/guidances/ucm128204.pdf Fung, D. Y. (2002). Rapid Methods and Automation in Microbiology. Comprehensive Reviews in Food

Science and Food Safety, 1.Hansen, S. L., Hardesty, D. R., & Myers, B. M. (1974). Evaluation of the BBL Minitek System for the

Identification of Enterobacteriaceae. Applied Microbiology, 28 (5), 798*801.Healthcare Commerce Media Corporation. (April 09, 2010). Today's global biopharma cold-chain market is

$5.1 billion, growing to $6.6 billion by 2011. Pharmaceutical Commerce Magazine.http://pharmaceuticalcommerce.com/index.php?pg=latest_news&articleid=1987

Hoerl, R., & Snee, R. D. (2012). Statistical thinking: improving business performance (2nd Ed.). Hoboken, NJ: John Wiley & Sons, Inc.

Hoizey, G., Lamiable, D., Robinet, A., Ludot, H., Malinovsky, J. M., Kaltenbach, M. L., . . . Millart, H.(2005). Sensitive bioassay of bupivacaine in human plasma by liquid-chromatography-ion trap massspectrometry. J Pharm Biomed Anal, 39(3-4), 587-592. doi: 10.1016/j.jpba.2005.03.042

International Conference on Harmonisation. (2003). Stability Testing of New Drug Substances and Products:

Q1A (R2) International Conference on Harmonisation of Technical Requirements for Registration ofPharmaceuticals for Human Use.

JUCE. (2014). The Deming Price. Retrieved 25 March, 2014, from http://www.juse.or.jp/e/deming/ Keller, H. (1982). Computers in the Modern clinical Laboratory: Interfacing and Sample Identification. Pure

and Applied Chemistry, 54(11), 2059-2062.Learning and Skills Improvement (LSIS) (Producer). (2013, 10 June 2013). Module 1, 2 and 3. Warehousing.

Retrieved from http://rwp.excellencegateway.org.uk/Embedded%20Learning/Vocational/Warehousing/ Lee, H. H., Dineva, M. A., Chua, Y. L., Ritchie, A., Ushiro-Lumb, I., & Wisniewski, C. A. (2010). Simple

amplification-based assay: A nucleic acid-based point-of-care platform for HIV-1 testing. Journal of

Infections Diseases 201(Supplement 1), S65-S71.Markens, U. (2009). Conducting Stability Studies: Recent Changes to Climatic Zone IV .Matthias, D. M., Robertson, J., Garrison, M. M., Newland, S., & Nelson, C. M. (2007). Freezing

temperatures in the vaccine cold chain: A systematic literature review. Vaccine, 25(20), 3980–3986.McGlennen, R. C. (2001). Miniaturization Technologies for Moelcular Diagnostics. Clinical Chemistry,

47 (3), 393-402.Montgomery, D. C. (2013). Design and analysis of experiments (8th Ed.). Hoboken, NJ John Wiley & Sons,

Inc. Nelson, C. M., Froes, P., VanDyck, A. M., Chavarría, J., Boda, E., Coca, A., . . . Lima, H. (2007). Monitoring

temperatures in the vaccine cold chain in Bolivia. Vaccine, 25(3), 433–437. Nelson, C. M., Wibisono, H., Purwanto, H., Mansyur, I., Moniaga, V., & Widjaya, A. (2004). Hepatitis B

vaccine freezing in the Indonesian cold chain: evidence and solutions. Bulletin of the World Health

Organization 82(2), 99-105. Novales-Li, P. (2013). Regulatory Perspectives on Companion Diagnostics. Regulatory Focus. Retrieved

9/11/13Parker, G. G., & Anderson.jr., E. G. (2002). Supply-chain integration: putting humpty dumpty back together

again. In T. Boone & R. Ganeshan (Eds.), New directions in supply-chain management: technology,strategy, and implementation (pp. 352-376). New York: AMACOM.

Petts, D. (2012). Letters of Consequence: 1912 - 2012 a History of the Institute of Biomedical Science:Institute of Biomedical Science.

Race, G. J., Tillery, G. W., & Dysert, P. A., 2nd. (2004). A history of pathology and laboratory medicine atBaylor University Medical Center. Proc (Bayl Univ Med Cent). 2004 Jan;17(1):42-55.

Ramanujam, R., Koelbl, J., & Ting, E. (1993). Room-temperature-Stable PCR Reagents. Genome Research,

3, 75-76.Ramsey, F. L., & Schafer, D. W. (1997). The statistical sleuth: a course in methods of data analysis.

Belmont, CA: Duxbury Press.

http://www.fda.gov/downloads/regulatoryinformation/guidances/ucm128204.pdf


http://pharmaceuticalcommerce.com/index.php?pg=latest_news&articleid=1987


http://www.juse.or.jp/e/deming/



http://rwp.excellencegateway.org.uk/Embedded%20Learning/Vocational/Warehousing/









Rao, A., Carr, L. P., Dambolena, I., Kopp, R. J., Martin, J., & Rafii, F. (1996). Total quality management: a

cross-functional perspective. New York, NY: John Wiley & Sons, Inc.Rodrigue, J.-P. (2013). The geography of transport systems (3rd ed.). New York: Routledge.Roy, R. K. (2010). A primer on the Taguchi method Dearborn, Mich. : Society of Manufacturing Engineers.Seevers, B. (2011). A Stability Budget as a Means of Protecting Drug Quality in the Distribution

Environment. An Update from the PDA PCCIG Task Force. PDA Cold Chain Conference March 12011. Seligson, D. (1966). Clinical laboratory automation. Journal of Chronic Diseases, 19(4), 509-517. doi:

http://dx.doi.org/10.1016/0021-9681(66)90124-X Shott, J. P., Galiwango, R. M., & Reynolds, S. J. (2012). A quality management approach to implementing

point-of-care technologies for HIV diagnosis and monitoring in Sub-Saharan Africa. Journal of

Tropical Medicine, Volume 2012, 1-8.Smith, A. (2005). History of the Agar Plate. Laboratory News. Retrieved 2/11/13, from

http://www.labnews.co.uk/features/history-of-the-agar-plate/ Snee, R. D. (1990). Statistical thinking and its contribution to total quality The American Statistician, 44,

116-121.Soeung, S., Grundy, J., Biggs, B., Boreland, M., Cane, J., Samnang, C., & Maynard, J. (2004). Management

systems response to improving immunization coverage in developing countries: a case study fromCambodia. Rural and Remote Health 4(263), 1-22.

Tang, Y.-W., Procop, G. W., & Persing, D. H. (1997). Molecular diagnostics of infectious diseases. Clinical

Chemistry, 43(11), 2021-2038.Temprecord™ International Ltd. (2010). Products Scientific. Retrieved 12 June 2013, from

http://www.temprecord.com/scientific.html Thakker, Y., & Woods, S. (1992). Storage of vaccines in the community: weak link in the cold chain? British

Medical Journal, 304(6829), 756–758.Walsh, D. P. (2013). The impact of the GDP guidelines on design, qualification and validation of packaging

systems. International Pharmaceutical Industry 5(4), 64-68.Wians, F. H. (2009). Clinical Laboratory Tests: Which, Why, and What Do The Results Mean? Lab

Medicine, 40(2), 105-113. doi: 10.1309/lm4o4l0hhutwwuddWild, C. J., & Pfannkuch, M. (1999). Statistical thinking in empirical enquiry. International Statistical

Review, 67 , 223-265.Wilson, P. F., Dell, L. D., & Anderson, G. F. (1993). Root cause analysis: a tool for Total Quality

Management . Milawukee, WI: ASQ Quality Press.World Health Organisation. (2009). Annex 2: Stability testing of active pharmaceutical ingredients and

finished pharmaceutical products (No. 953).Wu, A. (2006). A selected history and future of immunoassay development and applications in clinical

chemistry. Clinica Chiimica Acta, 369(2), 119-124.Zuckerman, D. M., Brown, P., & Nissen, S. E. (2011). Medical device recalls and the FDA approval process.

Archives of Internal Medicine, 171(11), 1006-1011.

http://dx.doi.org/10.1016/0021-9681(66)90124-X

http://dx.doi.org/10.1016/0021-9681(66)90124-X

http://www.labnews.co.uk/features/history-of-the-agar-plate/


http://www.temprecord.com/scientific.html




http://dx.doi.org/10.1016/0021-9681(66)90124-X

Laboratory Cold Chain

Documents