<<

35 (2017) 2127–2133

Contents lists available at ScienceDirect

Vaccine

journal homepage: www.elsevier.com/locate/vaccine

Review Is freezing in the vaccine cold chain an ongoing issue? A literature review q ⇑ Celina M. Hanson , Anupa M. George, Adama Sawadogo, Benjamin Schreiber

United Nations Children’s Fund (UNICEF), NY, NY, USA article info abstract

Keywords: Vaccine exposure to temperatures below recommended ranges in the cold chain may decrease vaccine Cold chain potency of freeze-sensitive leading to a loss of vaccine investments and potentially places chil- Supply chain dren at risk of contracting vaccine preventable illnesses. Vaccine freezing This literature review is an update to one previously published in 2007 (Matthias et al., 2007), analyz- Temperature monitoring ing the prevalence of vaccine exposure to temperatures below recommendations throughout various seg- Freeze-sensitive vaccine Training ments of the cold chain. Overall, 45 studies included in this review assess temperature monitoring, of which 29 specifically assess ‘too cold’ temperatures. The storage segments alone were evaluated in 41 articles, 15 articles examined the transport segment and 4 studied outreach sessions. The sample size of the studies varied, ranging from one to 103 shipments and from three to 440 storage units. Among reviewed articles, the percentage of vaccine exposure to temperatures below recommended ranges dur- ing storage was 33% in wealthier countries and 37.1% in lower income countries. Vaccine exposure to temperatures below recommended ranges occurred during shipments in 38% of studies from higher income countries and 19.3% in lower income countries. This review highlights continuing issues of vaccine exposure to temperatures below recommended ranges during various segments of the cold chain. Studies monitoring the number of events vaccines are exposed to ‘too cold’ temperatures as well as the duration of these events are needed. Many reviewed studies emphasize the lack of knowledge of health workers regarding freeze damage of vaccines and how this has an effect on temperature monitoring. It is important to address this issue by educating vaccina- tors and cold chain staff to improve temperature maintenance and supply chain management, which will facilitate the distribution of potent vaccines to children. Ó 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http:// creativecommons.org/licenses/by/4.0/).

Contents

1. Introduction ...... 2128 2. Methods ...... 2128 2.1. Identification of temperature monitoring studies ...... 2128 2.2. Data extraction ...... 2128 2.3. Analysis ...... 2128 3. Results...... 2128 4. Discussion...... 2130 5. Conclusion ...... 2131 Conflicts of interest...... 2132 Acknowledgements ...... 2132 References ...... 2132

q Open Access provided for this article by PATH. ⇑ Corresponding author. E-mail address: [email protected] (C.M. Hanson). http://dx.doi.org/10.1016/j.vaccine.2016.09.070 0264-410X/Ó 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). 2128 C.M. Hanson et al. / Vaccine 35 (2017) 2127–2133

1. Introduction 1. Year the study was published. 2. Year the study was conducted. Immunizations are hailed as one of the most important public 3. The country in which the study was conducted. health interventions known saving millions of lives every year. 4. Type of temperature measuring equipment used. Vaccines are responsible for the eradication of smallpox and the 5. Freeze threshold temperature or the temperature that was global community is currently working towards the eradication considered ‘‘too cold”. of polio. To reach current disease elimination and eradication tar- 6. Duration vaccines were at or below the freeze threshold (the gets, high immunization coverage rates are required. A well- time threshold after which was considered as ‘‘too cold”). functioning cold chain is at the center of ensuring potent vaccines 7. The duration of temperature monitoring. reach their intended population in an equitable and timely manner 8. The frequency of temperature monitoring. [2,3]. 9. Unit of analysis (, transport in vaccine carriers, Vaccines are biological products that slowly become inactive transport in refrigerated trucks, etc.). over time and must be kept within narrow temperature ranges 10. Study sample size (number of refrigerators or shipments). from manufacturers to those receiving them [3]. When exposed 11. Number of samples that registered freezing temperatures. to temperatures outside of this narrow range, the loss of potency 12. Percentage of samples that registered freezing temperatures may be accelerated [4]. Before 2007, attention had been focused or temperatures out of the recommended range of 2–8 °C. on examining the effects of vaccine exposure to heat. However, 13. Minimum temperatures observed. exposure to freezing temperatures are equally damaging to the quality of many vaccines as the process renders them inactive [4]. When exposed to freezing temperatures, the adjuvants con- 2.3. Analysis tained in some vaccines clump together adversely affecting the immunological properties of these vaccines [4]. Thus, the World Data were analyzed from studies that reported temperature Health Organization (WHO) recommends the following vaccines monitoring results. Different segments of the cold chain were ana- be stored and transported at 0–10 °C: Diptheria-tetanus contain- lyzed separately to account for differences in endpoints across ing vaccines, tetanus toxoid (TT), and B, human papil- studies. Analysis of data from published studies includes means lomavirus (HPV), meningitis C, pneumococcal (PCV), cholera, and standard deviations (SD) of study sample sizes and years of influenza, haemophilus influenza b (Hib), typhoid and inacti- studies as well as the weighted means and SDs of the percent of vated poliovirus (IPV) [4]. In 2015, approximately 59% of vacci- samples found below recommended temperatures. Some studies nes procured from the United Nations Children’s Fund provide temperature data for more than one administrative level (UNICEF) Supply Division’s (SD) catalogue are freeze-sensitive and for more than one cold chain segment; therefore, an individual [5]. Furthermore, the number of freeze-sensitive vaccines recom- study may have multiple data points. In addition, each country was mended by the WHO has increased by 50% over the last ten categorized by region and World Bank income status [7]. Wealthier years [4,6]. countries are defined as those with upper-middle to high income In 2007, a review of studies examining temperatures in the vac- status and lower income countries are defined as those in low to cine cold chain from 1985 to 2006 reported that in all segments of lower-middle income status. distribution, vaccines in 75–100% of monitored vaccine shipments Rigorous monitoring was defined as studies in which both the had been exposed to freezing temperatures [1]. This 2007 review frequency and duration of exposure to temperatures below recom- seems to have raised awareness about the risks of vaccine expo- mended temperatures were reported. For duration of monitoring, sure to freezing temperatures in both developed and developing anything more than 1 week was assigned a 1 and less was assigned countries of varying climates. a 0. For frequency of monitoring, any study with continuous mon- The objective of this updated review is to examine the results of itoring was given a 1, while spot checks or one time point only more recent monitoring studies regarding freezing temperatures measurements were given a 0. Rigor was calculated by then multi- within the immunization supply chain. plying the two to determine a single indication (0 or 1). Rigorous studies were identified by a 1. Data was entered, stored and tabulated in ExcelTM (Microsoft, 2. Methods Redmond, WA).

2.1. Identification of temperature monitoring studies 3. Results For this review, seven search terms were used alone and in combination to search gray and peer reviewed literature in The literature and web search was completed on March 21, PubMed, Popline, Embase, Biosis, and Google Scholar. In addition, 2016 and yielded 1070 articles. Screening of these articles led to the WHO and TechNet21 websites were searched for gray litera- the removal of 889 citations based on relevance of abstracts and ture. The search terms were (i) vaccine, (ii) temperature, (iii) ther- titles. The remaining 181 full text citations were screened accord- mostability, (iv) storage, (v) transport, (vi) freeze, and/or (vii) cold ing to the study inclusion criteria. The secondary and final screen chain. All searches were limited to studies published between July yielded 45 articles for final inclusion in the review (Tables 1 and 2). 2006 to August 2015, and did not include studies from Matthias Of the 45 articles, 33 are from peer-reviewed journals and 12 et al. [1]. In addition, unpublished studies that evaluated the tem- are from gray or unpublished literature. Thirteen articles describe perature of the storage of vaccines either in facilities, during trans- temperature monitoring data from the African region, 12 from portation or at outreach sessions in any country were included in the Western Pacific region, nine from the South East Asian region, this review. three from the Americas, four from the Eastern Mediterranean and four from Europe. 2.2. Data extraction Studies from all income levels are represented. Seven studies are from high-income countries (HIC), 13 are from upper-middle The following information was extracted from each of the stud- income countries (UMIC), 15 are from lower-middle income coun- ies that met the inclusion criteria: tries (LMIC) and eight are from low-income countries (LIC). Two Table 1 Vaccine cold chain studies reporting storage, transport, or immunization session temperatures below the recommended range included in both the literature review and the analysis.

No. Article Peer or gray Year of study Country Unit(s) of analysis Cold chain Study No. of shipments/ Percent (%) shipments/ segment sample size storage units found storage units found below threshold below threshold 1 Angoff et al. [25] Peer – USA Refrigerators and cold boxes Storage 45 40 89% 2 Arsalan et al. [17] Peer 2014 Pakistan Refrigerators Storage 379 6 1.60% 3 Azira et al. [15] Peer 2010 Malaysia Refrigerators Storage 89 24 27% 4 Carr et al. [26] Peer 2003–2004 Australia Refrigerators Storage 256 32 12.50% 5 de Timóteo Mavimbe et al. [27] Peer 2003 Mozambique Refrigerators Storage 9 3 33% 6 Govani and Sheth [28] Peer 2012–2013 India Refrigerators Storage 51 4 8% 7 The Government of the Philippines Gray 2010 Philippines Refrigerators Storage 16 8 50% and UNICEF [29] Cold boxes Transport 14 2 14% ..Hno ta./Vcie3 21)2127–2133 (2017) 35 Vaccine / al. et Hanson C.M. 8 Kartog˘lu et al. [30] Peer 2008–2009 Albania Refrigerators Storage 39 14 36% 9 Kartog˘lu et al. [18] Peer 2002–2004 Nepal, Myanmar, Cold boxes and Transport Nepal: 5 shipments; Nepal: 0 0 Turkey and Zimbabwe vaccine carriers Myanmar: 1 shipment Myanmar: 0 0 Turkey: 17 shipments Turkey: 0 0 Zimbabwe: 11 shipments Zimbabwe: 0 0 10 Levinson and Inspector General [31] Gray 2010 USA Refrigerators Storage 45 19 42% 11 McColloster and Vallbona [23] Peer – USA Refrigerators Storage 54 13 24% 12 Murhekar et al. [32] Peer 2012 India Refrigerators Storage 131 87 66% Vaccine carriers Transport 103 73 71% Vaccine carriers Outreach 40 2 5% 13 Özdemir and Özer [33] Peer __ Turkey Vaccine carriers Transport 7 0 0 14 PATH [34] Gray 2010 Albania Refrigerators Storage 21 1 4.8% 15 Ren et al. [35] Peer 2007 China Vaccine carriers Transport 7 7 100% 16 Robertson et al. [36] Gray 2009 Uganda Refrigerators Storage 24 20 83% Cold boxes Outreach 103 16 16% 17 Techathawat et al. [37] Peer 2004 Thailand Refrigerators Storage 96 31 32% Test boxes and vaccine carriers Transport 43 41 95% 18 UNICEF [38] Gray 2010–2011 Mongolia Refrigerators Storage 24 8 33% Study boxes Transport 18 6 33% 19 UNICEF [39] Gray 2014–2015 Cambodia Refrigerators Storage 19 9 47% Study boxes Transport 32 10 31% Study boxes Outreach 3 1 33% 20 Wang et al. [40] Peer 2003–2004 China Refrigerators Storage 3 2 33% Vaccine carriers Transport 3 0 0% 21 Yakum et al. [3] Peer 2013 Cameroon Refrigerators Storage 50 4 8% 2129 2130 C.M. Hanson et al. / Vaccine 35 (2017) 2127–2133

Table 2 Other articles included in the literature review.

No. Article Peer or gray Year of study Country Category 1 Alabachew [11] Gray 2014 Ethiopia Storage 2 Ameen et al. [12] Peer 2010 Nigeria Storage 3 Ateudjieu et al. [19] Peer 2008 Cameroon Storage 4 Breakwell et al. [21] Gray 2014 Micronesia, Federated States of Transport 5 Burstein et al. [41] Peer 2012 Ghana, Kenya, Uganda Storage 6 Chaudhri et al. [42] Gray 2010 Albania Storage 7 D’Onise et al. [20] Peer 2009 Australia Storage 8 Dubrava et al. [16] Peer 2006 Australia Storage/transport 9 El-Hamadi et al. [43] Peer 2014–2015 Libya Storage 10 Galhotra et al. [44] Peer 2006 India Storage/transport 11 Gopal-Krishnan et al. [45] Peer – Malaysia Storage 12 Gupta and Gupta [46] Peer 2015 India Storage 13 Karinagannanavar et al. [47] Peer 2010–2011 India Storage 14 Krishnappa et al. [48] Peer 2008 India Storage 15 Lloyd et al. [9] Peer 2011–2012 Tunisia Storage/transport 16 Lydon et al. [49] Peer 2005 and 2010 South Africa Storage/transport 17 Makuru [50] Gray 2012 Tanzania Storage 18 Mallik et al. [10] Peer 2008–2009 India Storage 19 Mugharbel and Al Wakeel [13] Peer 2007 Saudi Arabia Storage 20 Nexleaf Analytics and VillageReach [14] Gray 2014 Mozambique Storage 21 Pillay [51] Peer 2013 India Storage 22 Robertson et al. [8] Gray 2009 Vietnam Storage/transport/outreach sessions 23 Rogie et al. [52] Peer 2011–2012 Ethiopia Storage 24 Schlumberger et al. [53] Peer – Chad Storage

articles describe data from more than one country and these coun- tures below recommended ranges during outreach sessions was tries represent various income levels. 18% (Table 3). Studies defined the threshold for low, cold or freezing temper- Recommendations were included in all studies and the most atures at or below 2 °C in 37 studies; however, three studies commonly cited recommendation for countries are to conduct defined the threshold at or below 0 °C and five studies did not training sessions (19 articles), to improve supervision (10 articles) specifically mention a low threshold. The lowest temperature and the need for policies and guidelines (5 articles). reported was À21.3 °C and the highest reported was 31.7 °C. Com- plete temperature ranges were not available in 31 studies. Eigh- 4. Discussion teen studies evaluated storage units or shipments that fell out of ° the recommended range of 2–8 C, however, they did not specify Matthias et al. [1] identified 35 articles published between 1985 exact temperatures or whether freezing occurred. and June 2006 that assessed vaccine exposure to ’too cold’ temper- Assessment of the cold chain at a national level was performed atures, which equates to an average of 1.7 articles per year. This in seven studies, the provincial or regional level in 20 studies and update identified 21 articles in the last 10 years with data assess- the district level in 35 studies. One study did not specify an admin- ing vaccine exposure to temperatures below the recommended istrative level. range in vaccine cold chains, which is a rate of 2.1 articles on aver- All cold chain segments were represented in the final 45 arti- age per year. This may indicate a slight increase in awareness of cles. The storage segments alone were evaluated in 29 articles vaccine exposure to freezing temperatures. Over the last ten years, and four articles only examined the transport segment. Eight arti- improvements in cold chain monitoring have been reported due to cles studied storage and transport segments; three studies exam- newer cold chain technology and more studies are implementing ined storage, transport and outreach sessions and one article rigorous temperature monitoring techniques, as defined in this studied storage and outreach sessions. article, than were reported in 2007 [1]. In Vietnam, continuous The sample size of the studies varied, ranging from one to 103 temperature monitoring systems showed that vaccines in vaccine shipments or from three to 440 storage units. Storage units were carriers monitored during the study were not exposed to temper- described as walk-in-coolers, domestic refrigerators, -lined atures below 0 °C and enabled the country to identify temperature refrigerators, kerosene electric refrigerators, purpose built refriger- excursions during storage and outreach sessions [8]. Similarly, ators, bar-type refrigerators, cold boxes, vaccine carriers and study exposure to temperatures below recommended ranges during boxes. The transport segments and outreach sessions utilized vac- transport decreased by 12.1 percentage points and freeze alarms cine carriers or cold boxes. were reduced by 40% in Tunisia due to the use of continuous tem- Further analysis concentrating on data from studies that perature monitoring and Phase Change Material (PCM) transport assessed vaccine exposure to temperatures below recommended packs [9]. India reported significant improvements in the mainte- ranges included 21 articles (Table 1). Twenty-four studies were nance of cold chain temperatures after reorganizing the cold chain excluded based on a lack of information. This analysis found that and training health workers [10]. the percent of shipments found below recommended temperatures The same analysis completed by Matthias et al. [1] (Table 3) in wealthier countries was 38% and 19.3% in lower income coun- was also completed in this updated review. Since 2007, a greater tries. The percent of storage units where vaccines were exposed number of articles from low to lower-middle income countries to temperatures below recommended ranges were 33.3% in have completed studies examining the transport and outreach seg- wealthier countries and 37.1% in lower income countries. Three ments of the cold chain. temperature monitoring studies during outreach sessions were The occurrence of temperatures below the ‘too cold’ threshold found in lower income countries and none were found from in studies that monitored temperatures during transport showed wealthier countries. The percent of vaccine exposure to tempera- that shipments in lower income countries was 35.3% in the set of C.M. Hanson et al. / Vaccine 35 (2017) 2127–2133 2131

Table 3 Summary of studies.

Cold chain segment LMIC/LICb (Lower Income Countries) UMIC/HICc (Wealthier Countries) Transport Number of studies 6 6 Study sample size (mean (SDa)) 27.7 (38.4) 15.8 (14.6) Year study conducted (mean (SDa)) 2007 (5.6) 2005 (3.2) Rigorous monitoring (% of studies in group) 50% (3/6) 50% (3/6) % Shipments found below recommended threshold (weighted mean (SDa)) 19.3 (28.1) 38 (47.9) Storage Number of studies 8 10 Study sample size (mean (SDa)) 85 (125) 67 (72) Year study conducted (mean (SDa)) 2011 (3.6) 2007 (3.3) Rigorous monitoring (% of studies in group) 50% (4/8) 70% (7/10) % Storage units found below recommended threshold (weighted mean (SDa)) 37.1 (29.7) 33.3 (22.5) Outreach Number of studies 3 0 Study sample size (mean (SDa)) 48.7 (50.6) N/Ad Year study conducted (mean (SDa)) 2012 (2.5) N/A Rigorous monitoring (% of studies in group) 33% N/A % Storage units found below recommended threshold (weighted mean (SDa)) 18 (14.1) N/A

a Standard deviation. b LMIC/LIC = lower middle income countries/low income countries. c UMIC/HIC = upper middle income countries/high income countries. d N/A = not applicable.

studies presented by Matthias et al. (19.3% in the set analyzed Of utmost importance is the link between vaccine exposure to here) [1]. Shipments exposed to ‘too cold’ temperatures in wealth- temperatures below recommendations and disease outbreaks. In ier countries were 16.7% in the 2007 paper (38% here). However, a 20-year period, Micronesia did not have a single reported case the percent of rigorous studies, as defined here, conducted during of measles. However, in 2014 a relatively large measles outbreak transport was 50% for wealthier countries in 2007 (50% here) and occurred where more than half of all cases were vaccinated adults 78% for lower income countries (50% here Table 3). [21]. Micronesia’s cold chain had a history of cold chain breaches, In studies that monitored temperatures during , including freezing temperatures during transport, which point to the occurrence of exposure to temperatures considered ‘too cold’ in possible causes of vaccine failure [21]. Ambient freezing tempera- wealthier countries was 29% in 2007 (33.3% here), and 21.9% in tures during winter in Mongolia can be as low as À15 °C. Though lower income countries in 2007 (37.1% here). The percentage of the temperatures of vaccine storage units were not evaluated, chil- storage monitoring studies with rigorous methods, in lower dren vaccinated during winter months were over two times more income countries was 50% in 2007 (50% here). For wealthier coun- likely to become diagnosed with Hepatitis B implying the inactiva- tries, rigorous methods were found in 29% of storage monitoring tion of Hepatitis B vaccine due to exposure to freezing tempera- studies in 2007 (70% here Table 3). tures [22]. Finally, a study in the United States noted that health A possible reason for the differences in findings between the regions with higher percentages of refrigerators with frozen tem- two papers is the percentage of studies with rigorous methods. peratures also had higher rates of pertussis [23]. This is important since Matthias et al. found that rigorous studies Although this study presents current information on vaccine were a significant predictor of exposure to temperatures below temperature monitoring and vaccine exposure to temperatures recommended ranges. Since rigor, as defined here, takes into considered ‘too cold’, there are several limitations. Altogether, account continuous monitoring and the duration of monitoring, these studies represent a very small number of worldwide ship- the longer the duration, the more likely a ‘too cold’ event is to be ments, storage units and outreach sessions. Also, the number of discovered. Thus, for instance, the decrease in shipments in lower studies from upper-middle to high-income countries is similar to income countries observed with ‘too cold’ events between the two those from low to -lower-middle income countries; however, more studies may, in part, be due to the inclusion of studies with less studies are from the African (12) and West Pacific (12) regions. continuous monitoring in this study as compared to the 2007 arti- Thus, while these study findings may be indicative, it is difficult cle. These factors should be considered in the study design of to characterize them as representative of the global reality. future temperature monitoring studies. Studies were subject to a variety of biases. Selection bias may Recommendations ranged from technical guidance of vaccine have occurred in the selection of study sites or shipment routes. placement in storage units and cold chain product preferences to It is difficult to discern if, in some studies, sites or routes were cho- the development of local and national guidelines. The majority of sen due to past good performance, new equipment or ease of studies recognized the importance of engaging all segments of cold accessibility. chain staff by providing ongoing educational sessions and training sessions aimed at new staff members [11–14]. Supervision was another area of concern. Some articles mentioned the designation 5. Conclusion of one or two persons to supervise cold chain temperatures as a way to improve supervision and accountability [15,16]. Finally, This updated review highlights continuing issues of vaccine some articles reported a need for the development of local or exposure to ‘too cold’ temperatures during various segments of national guidelines [17,18] to have clear back up plans to reduce the cold chain. The number of rigorous studies available from vaccine loss in the event of electrical failure and a need for making countries has increased and plays an important role in the detec- current guidelines readily available to staff [19,20]. tion of ‘too cold’ temperatures. 2132 C.M. Hanson et al. / Vaccine 35 (2017) 2127–2133

Since 2007, the number of freeze-sensitive vaccines procured by [17] Arsalan Adeel, Naqvi Syed Baqar Shyum, Azfar Iqbal Osama Shakeel. UNICEF Supply Division (SD) for countries immunization programs Temperature monitoring of vaccines’ storage compartments in different health centres and pharmacies at Karachi, Pakistan. Int J 2014;5(3):984–8. increased by 50%. In 2015, freeze-sensitive vaccines procured [18] Kartog˘lu Ü et al. Use of cool water packs to prevent freezing during through UNICEF SD totaled a value of $1.2 billion USD [24], high- vaccine transportation at the country level. PDA J Pharm Sci Technol lighting the importance of monitoring vaccine exposure to ‘too 2009;63(1):11. [19] Ateudjieu J et al. Program on immunization and cold chain monitoring: cold’ temperatures. More rigorous monitoring studies are needed the status in eight health Districts in Cameroon. BMC Res Notes 2013;6(1):101. as well as more research to better understand instances of vaccine [20] D’Onise K et al. Have purpose-built vaccine refrigerators reduced the exposure to ‘too cold’ temperatures and the link to disease out- cost of vaccine losses in South Australia? Aust N Z J Public Health 2012;36 (6):572–6. breaks. As freezing remains a problem, the new categories of pre- [21] Breakwell Lucy et al. Measles outbreak associated with vaccine failure in qualification status for user independent freeze-free cold chain adults-federated states of Micronesia, February–August 2014. MMWR Morb equipment will be necessary. In addition, the emergence of actual Mortal Wkly Rep 2015;64(38):1088. [22] Dav aalkham D et al. Administration of hepatitis B vaccine in winter as a technologies on the market that meet freeze safety standards and significant predictor of the poor effectiveness of vaccination in rural Mongolia. uptake of these technologies by consumers will be essential. J Epidemiol Community Health 2007;61(7):578–84. Finally, until thermostable vaccines and freeze-free technologies [23] McColloster P, Vallbona C. Graphic-output temperature data loggers for become available in every country, countries can improve vaccine monitoring vaccine : implications for pertussis. Am J Public Health 2011;101(1):46–7. exposure to freezing by implementing continuous temperature [24] UNICEF. Supply annual report 2015. UNICEF supply division. [accessed supervision and the access and development of national and local July 14, 2016]. [25] Angoff R et al. Visual indicators on vaccine boxes as early warning tools to guidelines regarding vaccine storage and freezing vaccines. identify potential freeze damage. Infect Diseases Clin Pract 2015;23(4):184–9. [26] Carr Christine, Byles Julie, Durrheim David. Practice nurses best protect the vaccine cold chain in general practice. Austr J Adv Nurs 2009;27(2):35. Conflicts of interest [27] de Timóteo Mavimbe João Carlos, Bjune Gunnar. Cold chain management: knowledge and practices in primary health care facilities in Niassa, The authors declare no conflict of interest. Mozambique. Ethiop J Health Dev 2007;21(2):130–5. [28] Govani Kapil J, Sheth Jay K. Evaluation of temperature monitoring system of cold chain at all Urban Health Centres (UHCs) of Ahmedabad Municipal Acknowledgements Corporation (AMC) area. Healthline J Ind Assoc Prevent Soc Med 2015;6 (1):41–5. [29] The Government of the Philippines and UNICEF. Study on temperature We thank Alyssa Sharkey, UNICEF, and Aurelia Gasca, Unicef SD, monitoring of EPI vaccine; 2010. for the guidance provided for this report. This research is funded by [30] Kartog˘lu Ü, Erida N, Maire D. Improving temperature monitoring in the vaccine cold chain at the periphery: An intervention study using a 30-day PATH. Ò electronic temperature logger (Fridge-tag ). Vaccine 2010;28 (24):4065–72. References [31] Levinson Daniel R, Inspector General. Vaccines for children program: vulnerabilities in vaccine management. Department of Health and Human Services: Office of Inspector General; 2012. [1] Matthias DMD, Robertson J, Garrison MMM, Newland S, Nelson C. Freezing [32] Murhekar Manoj V et al. Frequent exposure to suboptimal temperatures in temperatures in the vaccine cold chain: a systematic literature review. Vaccine vaccine cold-chain system in India: results of temperature monitoring in 10 2007;25:3980–6. http://dx.doi.org/10.1016/j.vaccine.2007.02.052. states. Bull World Health Organ 2013;91(12):906–13. [2] World Health Organization. Global vaccine action plan 2011–2020; 2013. [33] Özdemir H, Özer AY. Investigating cold-chain system and efficacy of vaccines [3] Yakum Martin N et al. Factors associated with the exposure of vaccines to reaching the end user in Turkey and related regulations. FABAD J Pharm Sci adverse temperature conditions: the case of North West region, Cameroon. 2010;35:93–104. BMC Res Notes 2015;8(1):277. [34] PATH. Assessment of a remote alarm system for vaccine storage in Albania. [4] World Health Organization. WHO vaccine management handbook. Module Seattle: PATH; 2011. VMH-E2-01.1. How to monitor temperatures in the vaccine supply chain; [35] Ren Q et al. Evaluation of an outside-the-cold-chain vaccine delivery strategy 2015. [accessed 22 April 2016]. [36] Robertson J, Mugisha E, Kumakech E, LaMontagne D, Janmohamed A, [5] UNICEF. Supplies and logistics. [accessed on Kristensen D, et al. Monitoring ambient temperatures experienced during April 24 2016]. delivery of human papillomavirus vaccine in Uganda. Seattle: PATH; 2010. [6] World Health Organization, Milstien JB, et al. Temperature sensitivity of [37] Techathawat Sirirat et al. Exposure to heat and freezing in the vaccine cold vaccines; 2006. chain in Thailand. Vaccine 2007;25(7):1328–33. [7] World Bank Data. [accessed 18 April 2016]. [38] UNICEF. Temperature monitoring of EPI in Mongolia; 2011. [8] Robertson J, Vu H, Le N, et al. Cold chain temperature monitoring in Vietnam: [39] UNICEF Cambodia Country Office. Temperature monitoring study: Cambodia; monitoring ambient and cold chain temperatures during delivery of human 2014. papillomavirus vaccine. Seattle: PATH; 2010. [40] Wang L et al. Hepatitis B vaccination of newborn infants in rural China: [9] Lloyd John et al. Reducing the loss of vaccines from accidental freezing in the evaluation of a village-based, out-of-cold-chain delivery strategy. Bull World cold chain: The experience of continuous temperature monitoring in Tunisia. Health Organ 2007;85(9):688–94. Vaccine 2015;33(7):902–7. [41] Burstein Roy et al. Assessing vaccine cold chain storage quality: a cross- [10] Mallik S, Mandal PK, Chatterjee C, et al. Assessing cold chain status in a metro sectional study of health facilities in three African countries. Lancet 2013;381: city of India: an intervention study. African Health Sci 2011;11(1):128–33. S25. [11] Alabachew Demelash. Quality assessment of routine expanded programme of [42] Chaudhri Rohit, Borriello G, Anderson R. Pervasive computing technologies to immunization (EPI) for children under five in Debre Markos town, East Gojjam, monitor vaccine cold chains in developing countries. IEEE pervasive Amhara regional state, North West Ethiopia, 2014. Diss. AAU; 2014. computing. special issue on information and communication technologies [12] Ameen HA et al. Vaccine Storage and Handling Practices among routine for development; 2012. immunization service providers in a metropolitan city of North-Central [43] El-Hamadi Miluda Ragiab et al. Cold chain status and vaccination activities at Nigeria. J Commun Med Prim Health Care 2015;26(2):18–28. vaccination centers in Tripoli, Libya. Hospital 2015;5:10–6. [13] Mugharbel Kholood M, Al Wakeel Safa M. Evaluation of the availability of cold [44] Galhotra Abhiruchi et al. Surveillance of cold chain system during intensified chain tools and an assessment of health workers practice in Dammam. J Family pulse polio programme-2006 in Chandigarh. Ind J Pediat 2007;74(8):751–3. Commun Med 2009;16(3):83. [45] Gopal-Krishnan S, Sararaks S, Amir LE, Ibrahim MY, et al. Vaccine storage in [14] Nexleaf Analytics and VillageReach. ColdTrace year one: feedback from the private practice: a community trial. J Family Med Commun Health 2014;1 field; 2015. (2):1011. [15] Azira B, Norhayati MN, Norwati D. Optimal temperature of cold chain and its [46] Gupta Amarnath, Gupta Rekha. Study of cold chain practices at community associated factors among general practitioners in Kelantan, Malaysia. Int J health centers of Damoh District of Madhya Pradesh. Ntl J Commun Med Collab Res Intern Med Public Health 2014;6(6):168–79. 2015;6(4):528–32. [16] Dubrava Suzanne, Tait Paul A, Humphris Mary-Ann. Identification of potential [47] Karinagannanavar Aravind et al. A study of measles vaccine storage conditions and actual breaches in the cold-chain within a tertiary hospital. J Pharm Pract at Bellary District. Int J Curr Res Rev 2013;5(5):101. Res 2007;37(1):16–8. C.M. Hanson et al. / Vaccine 35 (2017) 2127–2133 2133

[48] Krishnappa Lalitha et al. Evaluation of cold chain practices in urban health [51] Pillay Shamla. A descriptive study into the cold chain management of centers of a metro city in India. Natl J Community Med 2014;5:288–92. childhood vaccines by nurses in primary health care clinics in the [49] Lydon P et al. Outsourcing vaccine logistics to the private sector: the evidence uMgungundlovu. District Diss; 2015. and lessons learned from the Western Cape Province in South-Africa. Vaccine [52] Rogie Bezunesh, Berhane Y, Bisrat F. Assessment of cold chain status for 2015;33(29):3429–34. immunization in central Ethiopia. Ethiop Med J 2013;51(Suppl 1):21–9. [50] Makuru Mohamed. Assessment of vaccines distribution system in public [53] Schlumberger M et al. Computerized temperature monitoring of the vaccine healthcare facilities in coast region, Tanzania. Diss Muhimbili University of cold chain in a tropical climate, Chad. Med Trop: Rev Corps Sante Colon Health and Allied Sciences; 2012. 2011;71(3):264–6.