A staff support programme for rural hospitals in Nepal

ProblemDistrict hospitals in Nepal struggle to provide essential services such as caesarean sections.ApproachRetention of health workers is critical to the delivery of long-term, quality health-care services. To promote retention and enhance performance in rural public hospitals, the Government of Nepal and the Nick Simons Institute progressively implemented a rural staff support programme in remote hospitals. After competitive selection for a compulsory-service scholarship and training, family practice doctors who could do basic surgery, orthopaedics and obstetrics were hired under a binding three-year contract in each participating hospital. Comfortable living quarters and an Internet connection were provided for the resident doctors; in-service training for all staff and capacity development for each hospital’s management committee were provided.

Local setting

Nepal’s mountainous landscape, poverty and inequitable rural/urban distribution of health workers pose barriers to adequate health care.

Relevant changes

Between 2011 and 2015 family practice doctors were maintained in all seven programme hospitals. All hospitals became providers of comprehensive emergency obstetric care and served more patients. Compared with hospitals not within the programme, deliveries increased significantly (203% versus 71% increase, respectively; P = 0.002). The programme recently expanded to 14 hospitals.

Lessons learnt

A package of human resource supports can improve the retention of doctors and the use of remote hospitals. Factors contributing to the success of this programme were compulsory-service scholarship, central personnel management, performance-based incentives and the provision of comfortable living quarters.     

Comprehensiveness of primary services in the care of infectious tuberculosis patients in Rawalpindi,Pakistan

Setting:

All tuberculosis (TB) diagnostic centres of Rawalpindi District, Pakistan, including five tertiary care hospitals and 16 rural health centres.

Objective:

To identify among sputum smear-positive patients registered during 2009 in the laboratory register those who had not been recorded in a treatment register, defined in the study as ‘initial loss to follow-up’.

Design:

A retrospective record review of routine TB data.

Results:

There were 16 145 suspects screened for TB and recorded in the laboratory registers. Of 1698 smear-positive patients identified in the laboratory registers, 101 (6%) could not be identified in the treatment registers. Eighty-six (10%) of 842 patients in tertiary care hospitals and 15/856 (2%) in rural health centres were not recorded (OR 6.4, 95%CI 3.6–11.6, P < 0.01).

Conclusion:

The study shows a significant association between type of health facility and initial loss to follow-up. In rural health centres, the proportion lost to follow-up is low, reflecting more efficient care than in tertiary care hospitals. Strategies are urgently needed to improve the registration and follow-up of smear-positive cases in tertiary care institutions.     

Depression and Pesticide Exposures among Private Pesticide Applicators Enrolled in the Agricultural Health Study

Background

We evaluated the relationship between diagnosed depression and pesticide exposure using information from private pesticide applicators enrolled in the Agricultural Health Study between 1993 and 1997 in Iowa and North Carolina.

Methods

There were 534 cases who self-reported a physician-diagnosed depression and 17,051 controls who reported never having been diagnosed with depression and did not feel depressed more than once a week in the past year. Lifetime pesticide exposure was categorized in three mutually exclusive groups: low (< 226 days, the reference group), intermediate (226–752 days), and high (> 752 days). Two additional measures represented acute high-intensity pesticide exposures: an unusually high pesticide exposure event (HPEE) and physician-diagnosed pesticide poisoning. Logistic regression analyses were performed relating pesticide exposure to depression.

Results

After adjusting for state, age, education, marital status, doctor visits, alcohol use, smoking, solvent exposure, not currently having crops or animals, and ever working a job off the farm, pesticide poisoning was more strongly associated with depression [odds ratio (OR) = 2.57; 95% confidence interval (CI), 1.74–3.79] than intermediate (OR = 1.07; 95% CI, 0.87–1.31) or high (OR = 1.11; 95% CI, 0.87–1.42) cumulative exposure or an HPEE (OR = 1.65; 95% CI, 1.33–2.05). In analysis of a subgroup without a history of acute poisoning, high cumulative exposure was significantly associated with depression (OR = 1.54; 95% CI, 1.16–2.04).

Conclusion

These findings suggest that both acute high-intensity and cumulative pesticide exposure may contribute to depression in pesticide applicators. Our study is unique in reporting that depression is also associated with chronic pesticide exposure in the absence of a physician-diagnosed poisoning.     

Vital Signs: Alcohol Poisoning Deaths — United States, 2010–2012

Background

Alcohol poisoning is typically caused by binge drinking at high intensity (i.e., consuming a very large amount of alcohol during an episode of binge drinking). Approximately 38 million U.S. adults report binge drinking an average of four times per month and consuming an average of eight drinks per episode.

Methods

CDC analyzed data for 2010–2012 from the National Vital Statistics System to assess average annual alcohol poisoning deaths and death rates (ICD-10 codes X45 and Y15; underlying cause of death) in the United States among persons aged ≥15 years, by sex, age group, race/ethnicity, and state.

Results

During 2010–2012, an annual average of 2,221 alcohol poisoning deaths (8.8 deaths per 1 million population) occurred among persons aged ≥15 years in the United States. Of those deaths, 1,681 (75.7%) involved adults aged 35–64 years, and 1,696 (76.4%) involved men. Although non-Hispanic whites accounted for the majority of alcohol poisoning deaths (67.5%; 1,500 deaths), the highest age-adjusted death rate was among American Indians/Alaska Natives (49.1 per 1 million). The age-adjusted rate of alcohol poisoning deaths in states ranged from 5.3 per 1 million in Alabama to 46.5 per 1 million in Alaska.

Conclusions

On average, six persons, mostly adult men, die from alcohol poisoning each day in the United States. Alcohol poisoning death rates vary substantially by state.

Implications for Public Health Practice

Evidence-based strategies for preventing excessive drinking (e.g., regulating alcohol outlet density and preventing illegal alcohol sales in retail settings) could reduce alcohol poisoning deaths by reducing the prevalence, frequency, and intensity of binge drinking.     

Long-Term Prospective Study of 6104 Survivors of Arsenic Poisoning During Infancy Due to Contaminated Milk Powder in 1955

Background

In 1955, an outbreak of arsenic poisoning caused by ingestion of arsenic-contaminated dry milk occurred in western Japan. We assessed the excess mortality among Japanese who were poisoned during this episode as infants.

Methods

We identified and enrolled 6104 survivors (mean age at enrollment, 27.4 years) who had ingested contaminated milk when they were age 2 years or younger; they were followed until 2006 (mean duration of follow-up, 24.3 years). Death certificates of subjects who died between 1982 and 2006 were examined to calculate cause-specific standardized mortality ratios (SMRs) using the mortality rate among Osaka residents as the standard.

Results

There was no significant excess overall mortality (SMR: 1.1, 95% confidence interval: 1.0–1.2). However, significant excess mortality in both sexes was observed from diseases of the nervous system (3.7, 1.9–6.2). Excess mortality from all causes of death decreased to unity beyond 10 years after study enrollment. The 408 men who were unemployed at the time of enrollment in the study had a significantly elevated risk of death from diseases of the nervous system (25.3, 10.8–58.8), respiratory diseases (8.6, 3.1–16.8), circulatory diseases (3.2, 1.6–5.2), and external causes (2.6, 1.4–4.1).

Conclusions

As compared with the general population, survivors of arsenic poisoning during infancy had a significantly higher mortality risk from diseases of the nervous system.Key words: arsenic poisoning, mortality, prospective study, food poisoning, standardized mortality ratio     

Medical and nursing students' intentions to work abroad or in rural areas: a cross-sectional survey in Asia and Africa

Objective

To assess medical and nursing students’ intentions to migrate abroad or practice in rural areas.

Methods

We surveyed 3199 first- and final-year medical and nursing students at 16 premier government institutions in Bangladesh, Ethiopia, India, Kenya, Malawi, Nepal, the United Republic of Tanzania and Zambia. The survey contained questions to identify factors that could predict students’ intentions to migrate. Primary outcomes were the likelihoods of migrating to work abroad or working in rural areas in the country of training within five years post-training. We assessed predictors of migration intentions using multivariable proportional odds models.

Findings

Among respondents, 28% (870/3156) expected to migrate abroad, while only 18% (575/3158) anticipated a rural career. More nursing than medical students desired professions abroad (odds ratio, OR: 1.76; 95% confidence interval, CI: 1.25–2.48). Career desires before matriculation correlated with current intentions for international (OR: 4.49; 95% CI: 3.21–6.29) and rural (OR: 4.84; 95% CI: 3.52–6.66) careers. Time spent in rural areas before matriculation predicted the preference for a rural career (20 versus 0 years: OR: 1.53, 95% CI: 1.19–1.98) and against work abroad (20 versus 0 years: OR: 0.69, 95% CI: 0.50–0.96).

Conclusion

A significant proportion of students surveyed still intend to work abroad or in cities after training. These intentions could be identified even before matriculation. Admissions standards that account for years spent in rural areas could promote greater graduate retention in the country of training and in rural areas.     

Diarrhoea-related hospitalizations in children before and after implementation of monovalent rotavirus vaccination in Mexico

Objective

To assess, by socioeconomic setting, the effect of nationwide vaccination against species A rotavirus (RVA) on childhood diarrhoea-related hospitalizations in Mexico.

Methods

Data on children younger than 5 years who were hospitalized for diarrhoea in health ministry hospitals between 1 January 2003 and 31 December 2011 were collected from monthly discharge reports. Human development indexes were used to categorize the states where hospitals were located as having generally high, intermediate or low socioeconomic status. Annual rates of hospitalization for diarrhoea – per 10 000 hospitalizations for any cause – were calculated. Administrative data were used to estimate vaccine coverage.

Findings

In the states with high, intermediate and low socioeconomic status, coverage with a two-dose monovalent RVA vaccine – among children younger than 5 years – had reached 93%, 86% and 71%, respectively, by 2010. The corresponding median annual rates of hospitalization for diarrhoea – per 10 000 admissions – fell from 1001, 834 and 1033 in the “prevaccine” period of 2003–2006, to 597, 497 and 705 in the “postvaccine” period from 2008 to 2011, respectively. These decreases correspond to rate reductions of 40% (95% confidence interval, CI: 38–43), 41% (95% CI: 38–43) and 32% (95% CI: 29–34), respectively. Nationwide, RVA vaccination appeared to have averted approximately 16 500 hospitalizations for childhood diarrhoea in each year of the postvaccine period.

Conclusion

Monovalent RVA vaccination has substantially reduced childhood diarrhoea-related hospitalizations for four continuous years in discretely different socioeconomic populations across Mexico.     

DDT Exposure of Zebrafish Embryos Enhances Seizure Susceptibility: Relationship to Fetal p,p′-DDE Burden and Domoic Acid Exposure of California Sea Lions

Background

California sea lions have a large body burden of organochlorine pesticides, and over the last decade they have also been subject to domoic acid poisoning. Domoic acid poisoning, previously recognized in adult animals, is now viewed as a major cause of prenatal mortality. The appearance of a chronic juvenile domoic acid disease in the sea lions, characterized by behavioral abnormalities and epilepsy, is consistent with early life poisoning and may be potentiated by organochlorine burden.

Objective

We investigated the interactive effect of DDT (dichlorodiphenyltrichloroethane) on neurodevelopment using a zebrafish (Danio rerio) model for seizure behavior to examine the susceptibility to domoic acid–induced seizures after completion of neurodevelopment.

Methods

Embryos were exposed (6–30 hr postfertilization) to either o,p′-DDT or p,p′-DDE (dichlorodiphenyldichloroethylene) during neurodevelopment via a 0.1% dimethyl sulfoxide solution. These larval (7 days postfertilization) fish were then exposed to either the seizure-inducing drug pentylenetetrazol (PTZ) or domoic acid; resulting seizure behavior was monitored and analyzed for changes using cameras and behavioral tracking software.

Results

Embryonic exposure to DDTs enhanced PTZ seizures and caused distinct and increased seizure behaviors to domoic acid, most notably a type of head-shaking behavior.

Conclusion

These studies demonstrate that embryonic exposure to DDTs leads to asymptomatic animals at completion of neurodevelopment with greater sensitivity to domoic acid–induced seizures. The body burden levels of p,p′-DDE are close to the range recently found in fetal California sea lions and suggest a potential interactive effect of p,p′-DDE embryonic poisoning and domoic acid toxicity.     

Inequities in postnatal care in low- and middle-income countries: a systematic review and meta-analysis

Objective

To assess the socioeconomic, geographical and demographic inequities in the use of postnatal health-care services in low- and middle-income countries.

Methods

We searched Medline, Embase and Cochrane Central databases and grey literature for experimental, quasi-experimental and observational studies that had been conducted in low- and middle-income countries. We summarized the relevant studies qualitatively and performed meta-analyses of the use of postnatal care services according to selected indicators of socioeconomic status and residence in an urban or rural setting.

Findings

A total of 36 studies were included in the narrative synthesis and 10 of them were used for the meta-analyses. Compared with women in the lowest quintile of socioeconomic status, the pooled odds ratios for use of postnatal care by women in the second, third, fourth and fifth quintiles were: 1.14 (95% confidence interval, CI : 0.96–1.34), 1.32 (95% CI: 1.12–1.55), 1.60 (95% CI: 1.30–1.98) and 2.27 (95% CI: 1.75–2.93) respectively. Compared to women living in rural settings, the pooled odds ratio for the use of postnatal care by women living in urban settings was 1.36 (95% CI: 1.01–1.81). A qualitative assessment of the relevant published data also indicated that use of postnatal care services increased with increasing level of education.

Conclusion

In low- and middle-income countries, use of postnatal care services remains highly inequitable and varies markedly with socioeconomic status and between urban and rural residents.     

NHS Scotland reduces the postcode lottery for hip arthroplasty: an ecological study of the impact of waiting time initiatives

Objectives

Following the election of the Labour government in 1997, policies were developed in England to reduce waiting times for NHS treatment with commitments to reduce health inequalities. Similar policies were adopted in Scotland but with less emphasis on the use of the private sector to deliver NHS treatments than in England. This study uses routinely collected NHS Scotland data to analyse geographical and socioeconomic inequalities in elective hip arthroplasty treatment before and after the introduction of the waiting time initiatives in Scotland in 2003.

Design

Ecological study design.

Setting

Scotland.

Participants

NHS-funded patients receiving elective hip arthroplasty delivered by the NHS and private hospitals between 1 April 1998 and 31 March 2008.

Main outcome measures

Directly standardised treatment rates and incidence rate ratios calculated using Poisson regression.

Results

There was a 42% increase in NHS-funded hip arthroplasties carried out in Scotland from 4095 in 2002–2003 (1 April 2002–31 March 2003) to 5829 in 2007–2008. There is evidence of a statistically significant reduction in geographical inequality (likelihood ratio test p < 0.001) but no evidence of any statistically significant change in socioeconomic inequality (p = 0.108), comparing the 5 years after 1 April 2003 with the 5 years before 1 April 2003. The number of NHS-funded hip arthroplasties undertaken privately rose from 1.1% in 2002–2003 to 2.9% in 2007–2008, whereas the NHS Golden Jubilee National Hospital increased its share from 3.3% to 10.6% over the same period.

Conclusions

The reduction in geographical inequality, or ‘postcode lottery’, in hip arthroplasty treatment in Scotland may be due to increased NHS capacity, in particular the development of the NHS Golden Jubilee National Hospital in Clydebank, Greater Glasgow as a dedicated centre to reduce surgery waiting times.     

Mobilizing communities to improve maternal health: results of an intervention in rural Zambia

Objective

To determine whether a complex community intervention in rural Zambia improved understanding of maternal health and increased use of maternal health-care services.

Methods

The intervention took place in six rural districts selected by the Zambian Ministry of Health. It involved community discussions on safe pregnancy and delivery led by trained volunteers and the provision of emergency transport. Volunteers worked through existing government-established Safe Motherhood Action Groups. Maternal health indicators at baseline were obtained from women in intervention (n = 1775) and control districts (n = 1630). The intervention’s effect on these indicators was assessed using a quasi-experimental difference-in-difference approach that involved propensity score matching and adjustment for confounders such as education, wealth, parity, age and distance to a health-care facility.

Findings

The difference-in-difference comparison showed the intervention to be associated with significant increases in maternal health indicators: 14–16% in the number of women who knew when to seek antenatal care; 10–15% in the number who knew three obstetric danger signs; 12–19% in those who used emergency transport; 22–24% in deliveries involving a skilled birth attendant; and 16–21% in deliveries in a health-care facility. The volunteer drop-out rate was low. The estimated incremental cost per additional delivery involving a skilled birth attendant was around 54 United States dollars, comparable to that of other demand-side interventions in developing countries.

Conclusion

The community intervention was associated with significant improvements in women’s knowledge of antenatal care and obstetric danger signs, use of emergency transport and deliveries involving skilled birth attendants.     

Trends in socioeconomic disparities in a rapid under-five mortality transition: a longitudinal study in the United Republic of Tanzania

Objective

To explore trends in socioeconomic disparities and under-five mortality rates in rural parts of the United Republic of Tanzania between 2000 and 2011.

Methods

We used longitudinal data on births, deaths, migrations, maternal educational attainment and household characteristics from the Ifakara and Rufiji health and demographic surveillance systems. We estimated hazard ratios (HR) for associations between mortality and maternal educational attainment or relative household wealth, using Cox hazard regression models.

Findings

The under-five mortality rate declined in Ifakara from 132.7 deaths per 1000 live births (95% confidence interval, CI: 119.3–147.4) in 2000 to 66.2 (95% CI: 59.0–74.3) in 2011 and in Rufiji from 118.4 deaths per 1000 live births (95% CI: 107.1–130.7) in 2000 to 76.2 (95% CI: 66.7–86.9) in 2011. Combining both sites, in 2000–2001, the risk of dying for children of uneducated mothers was 1.44 (95% CI: 1.08–1.92) higher than for children of mothers who had received education beyond primary school and in 2010–2011, the HR was 1.18 (95% CI: 0.90–1.55). In contrast, mortality disparities between richest and poorest quintiles worsened in Rufiji, from 1.20 (95% CI: 0.99–1.47) in 2000–2001 to 1.48 (95% CI: 1.15–1.89) in 2010–2011, while in Ifakara, disparities narrowed from 1.30 (95% CI: 1.09–1.55) to 1.15 (95% CI: 0.95–1.39) in the same period.

Conclusion

While childhood survival has improved, mortality disparities still persist, suggesting a need for policies and programmes that both reduce child mortality and address socioeconomic disparities.     

Estimation of HIV Seroprevalence in Colorectal Hospitals by Questionnaire Survey in Korea, 2002–2007

Objectives

The incidence of anal disease is higher among persons with human immunodeficiency virus (HIV) infection than among the general population. We surveyed the status of seroprevalence in colorectal hospitals in Korea.

Methods

The survey was conducted in colorectal hospitals in Korea from November to December 2008. The questionnaire was comprised of six topics about the status of HIV testing in colorectal hospitals. We gathered the data by website (http://hivqa.nih.go.kr/risk) or fax.

Results

Among 774 colorectal hospitals contacted, 109 (14%) hospitals participated in the survey. Among these, 48 hospitals (44%) performed HIV tests in their own hospitals and 11 (23%) took HIV testing by rapid method. The main reason for recommending an HIV test was surgical operation (54%) followed by endoscope (11%) and health checkup (9%). The annual number of HIV tests increased from 58,647 (at 21 hospitals) in 2002 to 246,709 (at 58 hospitals) in 2007. HIV seroprevalence was >3.0 per 10,000 individuals during 2002–2005, decreased to 2.2 per 10,000 individuals in 2006 and rose to 2.8 per 10,000 individuals in 2007.

Conclusions

HIV seroprevalence of colorectal hospitals was more than twice that of general hospitals in Korea. HIV surveillance systems based on colorectal hospitals for HIV/AIDS transmission prevention by early HIV diagnosis are needed.     

Risk of self-harm and suicide in people with specific psychiatric and physical disorders: comparisons between disorders using English national record linkage

Background

Psychiatric illnesses are known risk factors for self-harm but associations between self-harm and physical illnesses are less well established. We aimed to stratify selected chronic physical and psychiatric illnesses according to their relative risk of self-harm.

Design

Retrospective cohort studies using a linked dataset of Hospital Episode Statistics (HES) for 1999–2011.

Participants

Individuals with selected psychiatric or physical conditions were compared with a reference cohort constructed from patients admitted for a variety of other conditions and procedures.

Setting

All admissions and day cases in National Health Service (NHS) hospitals in England.

Main outcome measures

Hospital episodes of self-harm. Rate ratios (RRs) were derived by comparing admission for self-harm between cohorts.

Results

The psychiatric illnesses studied (depression, bipolar disorder, alcohol abuse, anxiety disorders, eating disorders, schizophrenia and substance abuse) all had very high RRs (> 5) for self-harm. Of the physical illnesses studied, an increased risk of self-harm was associated with epilepsy (RR = 2.9, 95% confidence interval [CI] 2.8–2.9), asthma (1.8, 1.8–1.9), migraine (1.8, 1.7–1.8), psoriasis (1.6, 1.5–1.7), diabetes mellitus (1.6, 1.5–1.6), eczema (1.4, 1.3–1.5) and inflammatory polyarthropathies (1.4, 1.3–1.4). RRs were significantly low for cancers (0.95, 0.93–0.97), congenital heart disease (0.9, 0.8–0.9), ulcerative colitis (0.8, 0.7–0.8), sickle cell anaemia (0.7, 0.6–0.8) and Down''s syndrome (0.1, 0.1–0.2).

Conclusions

Psychiatric illnesses carry a greatly increased risk of self-harm as well as of suicide. Many chronic physical illnesses are also associated with an increased risk of both self-harm and suicide. Identifying those at risk will allow provision of appropriate monitoring and support.     

The impact of rural health system reform on hospitalization rates in the Islamic Republic of Iran: an interrupted time series

Objective

To assess the effects on hospital utilization rates of a major health system reform – a family physician programme and a social protection scheme – undertaken in rural areas of the Islamic Republic of Iran in 2005.

Methods

A “tracer” province that was not a patient referral hub was selected for the collection of monthly hospitalization data over a period of about 10 years, beginning two years before the rural health system reform (the “intervention”) began. An interrupted time series analysis was conducted and segmented regression analysis was used to assess the immediate and gradual effects of the intervention on hospitalization rates in an intervention group composed of rural residents and a comparison group composed of urban residents primarily.

Findings

Before the intervention, the hospitalization rate in the rural population was significantly lower than in the comparison group. Although there was no significant increase or decline in hospitalization rates in the intervention or comparison group before the intervention, after the intervention a significant increase in the hospitalization rate – of 4.6 hospitalizations per 100 000 insured persons per month on average – was noted in the intervention group (P < 0.001). The monthly increase in the hospitalization rate continued for over a year and stabilized thereafter. No increase in the hospitalization rate was observed in the comparison group.

Conclusion

The primary health-care programme instituted as part of the health system reform process has increased access to hospital care in a population that formerly underutilized hospital services. It has not reduced hospitalizations or hospitalization-related expenditure.     

Health-care seeking among people with cough of 2 weeks or more in India. Is passive TB case finding sufficient?

Setting:

Thirty districts of India.

Objectives:

To estimate 1) the proportion of people with cough of ≥2 weeks, 2) those who did not seek care from a health care provider for cough, and 3) their characteristics.

Methodology:

A cross-sectional community-based survey in which 4562 people (aged ≥18 years) were interviewed.

Results:

Of the 4562 individuals interviewed, 437 (9.5%, 95%CI 7.2–11.8) had cough ≥2 weeks; this was more frequent in those >55 years of age (14%) and in those from districts in eastern (12%) and northern (11%) states of India. Of those with cough, 300 (69%, 95%CI 60–77) had not sought care from any health care provider. Not seeking care was more frequent in people residing in rural areas (73%) compared with urban areas (53%), and in the districts of eastern (82%) and northern (74%) states compared to districts from the southern (46%) and western (54%) states.

Conclusions:

Nearly a tenth of those interviewed aged ≥18 years had cough of ≥2 weeks. About two thirds, especially those from rural areas, had not visited a health care provider for the cough. This finding has huge implications for India’s current mostly passive case-finding strategy for detecting and controlling tuberculosis.     

Hospital Surgical Volume and 3-Year Mortality in Severe Prognosis Cancers: A Population-Based Study Using Cancer Registry Data

BackgroundThe impact of hospital surgical volume on long-term mortality has not been well assessed in Japan, especially for esophageal, biliary tract, and pancreatic cancer, although these three cancers need a high level of medical-technical skill. The purpose of this study was to examine associations between hospital surgical volume and 3-year mortality for these severe-prognosis cancer patients.MethodsPatients who received curative surgery for esophageal, biliary tract, and pancreatic cancers were analyzed using the Osaka Cancer Registry data from 2006–2013. Hospital surgical volume was categorized into tertiles (high/middle/low) according to the average annual number of curative surgeries per hospital for each cancer. Three-year survivals were calculated using the Kaplan-Meier method. Hazard ratios (HRs) of 3-year mortality were calculated using Cox proportional hazard models, adjusting for patient characteristics.ResultsThree-year survival was higher with increased hospital surgical volume for all three cancers, but the relative importance of volume varied across sites. After adjustment for all confounding factors, HRs in middle- and low-volume hospitals were 1.34 (95% confidence interval [CI], 1.14–1.58) and 1.57 (95% CI, 1.33–1.86) for esophageal cancer; 1.39 (95% CI, 1.15–1.67) and 1.57 (95% CI, 1.30–1.89) for biliary tract cancer; 1.38 (95% CI, 1.16–1.63) and 1.90 (95% CI, 1.60–2.25) for pancreatic cancer, respectively. In particular for localized pancreatic cancer, the impact of hospital surgical volume on 3-year mortality was strong (HR 2.66; 95% CI, 1.61–4.38).ConclusionWe suggest that patients who require curative surgery for esophageal, biliary tract, and pancreatic cancer may benefit from referral to high-volume hospitals.Key words: 3-year mortality, biliary tract cancer, esophageal cancer, hospital surgical volume, pancreatic cancer