October 6, 1997 Please reply to: Frances Pritchard

Telephone No. 392-7025

Please refer to: 97hlh 9-11a.let

To: Medical Officer of Health for the New City of Toronto

Medical Officers of Health in the Province of Ontario

Council endorsed the following action taken by the Board of Health:

1. Adopted the report (September 2, 1997) from the Medical Officer of Health and forwarded same to the Toronto Transition Team, the Board of Health of the new City of Toronto and the new City of Toronto Council with the recommendation that the Healthiest Babies Possible program, be expanded to meet the need for low birthweight prevention in the new City of Toronto.

2. Distributed the report to all health units in Ontario, the Ministry of Health, the Ministry of Community and Social Services and the Office of Integrated Services.

3. Recommended program strategies be put in place to enable women to enter the program earlier in their pregnancies, and to maintain their participation to the end of their pregnancies as much as possible.

Yours truly,

Assistant City Clerk

_____la

FMP

enc.

Subject: The Effectiveness of Health Professional Home Visits and Milk Coupons in the Prevention of Low Birthweight: A Cohort Study of the City of Toronto Healthiest Babies Possible Program

Origin: Medical Officer of Health, September 2, 1997 (p:\1997\ug\cms\hld\hl970068.hld)

Recommendations:

1. That this report be adopted by the Board of Health and forwarded to the Toronto Transition Team, the Board of Health of the new City of Toronto and the new City of Toronto Council with the recommendation that the Healthiest Babies Possible program, be expanded to meet the need for low birthweight prevention in the new City of Toronto.

2. That this report be distributed to all health units in Ontario, the Ministry of Health, the Ministry of Community and Social Services and the Office of Integrated Services.

3. That program strategies be put in place to enable women to enter the program earlier in their pregnancies, and to maintain their participation to the end of their pregnancies as much as possible.

Comments:

The Healthiest Babies Possible (HBP) program has been in operation in the City of Toronto since 1979, providing in-home counselling and milk coupons to low income pregnant women with at least one additional risk factor for poor pregnancy outcome. The overall goal is to reduce the rate of low birthweight (LBW), babies less than 2500 grams, which is associated with developmental and medical problems at birth and throughout life. Public health dietitians and nurses work as a team to address the problems of undernutrition, pregnancy complications, medical conditions, substance use, stress, depression and social isolation among program participants.

The attached study is based on HBP program data collected from June 1994 to June 1996. It compares the outcomes of two groups of clients - those who entered the program late in pregnancy (Group A), and therefore received only a few home visits, with those who entered early in pregnancy (Group B), thus receiving more visits. Only clients who completed the program are included in this analysis.

Attached to this study is a comprehensive Program Evaluation of all HBP clients during the same period, providing more detailed information about the demographic profile, changes made and health outcomes of all program participants.

This study shows that the Healthiest Babies Possible (HBP ) program is effective in reducing the low birthweight rate among high risk, low income pregnant women and is associated with improvements in diet, smoking and depression. Positive outcomes correlate strongly with number of home visits. Among HBP clients who received only one or two visits, 27.7% had LBW babies, whereas among those who received more than eight visits, only 5.1% were LBW. The low birthweight rate for this latter group of high risk clients was similar to the rate for the city as a whole.

Also accompanying this report is a comprehensive program evaluation of all HBP clients during the same period providing more detailed information about the demographic profile and health outcomes of all program participants.

The information in this program evaluation demonstrates that the HBP program is highly effective in reaching an extremely high risk client group. Among other behavioural, medical and social improvements, the evaluation shows that breastfeeding at delivery has risen to 88% of clients, with enormous benefits to infants postnatally.

The Healthiest Babies Possible program is complementary to the new provincial Healthy Babies, Healthy Children program, which will provide funds for home visitors (mostly postnatal but ideally also prenatally). Health units will administer this program and facilitate appropriate referral. In this program, home visitors will be trained to refer high risk individuals to public health professionals for further assistance, such as HBP.

Conclusions:

Home counselling by public health dietitians and nurses, with provision of milk coupons, is an effective strategy for reaching high risk pregnant women; improving health, behavioural and socioenvironmental conditions; and reducing the low birthweight rate for this population. The optimal number of visits varies with each individual, but it appears that a low birthweight rate similar to the city-wide rate can be achieved with a minimum of eight visits.

Given the human and economic costs of low birthweight, an effective intervention to prevent low birthweight such as HBP should be available to every at-risk woman in the new City of Toronto.

(The Effectiveness of Health Professional Home Visits and Milk Coupons in the Prevention of Low Birthweight: A Cohort Study of the City of Toronto Healthiest Babies Possible Program prepared by the City of Toronto Department of Public Health, August 1997)

Acknowledgement:

The principal authors of this study are Ellen Desjardins, Heathiest Babies Possible Coordinator, and Deborah Hardwick, Epidemiologist, City of Toronto Public Health.

INTRODUCTION

This report summarizes the results of a study conducted among the Healthiest Babies Possible clients over a two year period from June 1994 to June 1996. We hypothesized that clients who received a greater number of visits from public health dietitians and nurses would show greater improvements in selected risks and conditions and lower incidence of low birthweight than clients with fewer visits.

The prevention of low birthweight (less than 2500 grams) is one of the most important concerns of perinatal care. Low birthweight (LBW) accounts for about three-quarters of early neonatal mortality in Canada and contributes significantly to infant and childhood morbidity (1,2,3). While the exact causes of LBW are not clearly known, the vast literature in this area indicates that a complexity of factors are involved and that mothers who are socially, educationally and economically disadvantaged have greater numbers of LBW infants.

The cost of caring for babies who weigh less than 2500 grams at birth is high, estimated at $500 to $1000 per day in Canada, without taking into consideration the emotional and financial long-term costs for families, the health care system and society at large (1,2,3). One study estimated that the total cost of caring for a very LBW baby (less than 1500 grams) until the age of two years is $200,000. For those infants who also face lifetime disability, these costs are likely to escalate considerably (3).

Healthiest Babies Possible (HBP) is a prenatal prevention program of Toronto Public Health that targets high risk women in an attempt to reduce the number of LBW babies and to improve other perinatal outcomes, through professional home visits by dietitians and public health nurses. Although public health nursing has undergone a shift in emphasis away from home visiting, there is some evidence that home visits are successful and cost efficient in certain cases. In a systematic literature review of home visiting, Ciliska et al. (4) found evidence that home visiting results in greater client use of community services and other social support networks, better maternal health outcomes, reduced levels of smoking and other factors associated with LBW. The effects of prenatal home visiting intervention are found to be greater among clients with the greatest number of risk factors, indicating that those pregnant women at highest risk for poor birth outcomes receive the greatest payoffs from professional home visiting. The authors also noted that most published reports on prenatal home visiting were not methodologically strong enough to show a difference in LBW because sample sizes were too small.

The strategy of the HBP program in Toronto involves:

1. reaching and maintaining contact with pregnant women at high risk for low birthweight,

2. understanding each woman's individual situation, both through formal assessment and more informal on-going communication,

3. facilitating changes which result in improved health of both the mother and her baby,

4. providing milk coupons for about one litre of milk a day.

Figure 1 illustrates the complexity of addressing low birthweight, and the risk factors and conditions that have impact on LBW.

Conceptual model for Healthiest Babies Possible program strategies

METHODS

Research Design

This is a prospective cohort study with no control group. The cohort is every client who entered the HBP program between June 1, 1994 and June 30, 1996 and stayed in the program until the birth of her baby.

A total of 419 women completed the program, representing 78% of the 539 women who entered. The most common reason for not completing the program was because the client moved out of city limits.

During the first prenatal visit by both a public health nurse and dietitian, each client is professionally assessed for a spectrum of risk factors. A comparison of clients who did and did not complete the program revealed that out of 24 risks initially assessed, only four risks--adolescence, lack of social support, abuse and smoking--were more prevalent among those who did not complete the program. Only singleton births were included in the research questions relating to low birthweight because of the strong relationship between multiple births and low birthweight, reducing the cohort size to 413 for most of the analysis.

Definition of Treatment Groups

The major independent variable was exposure to the HBP program intervention (number of visits). The exposure to the duration of intervention has an upper limit, i.e. when the baby is born. However, mothers can enter HBP at any time during their pregnancy. Women entered as early as 4 weeks gestation and as late as 29 weeks, with an average registration at 16 weeks gestation. The number of visits received by clients is moderately to strongly correlated with gestational age at registration. Dietitians and public health nurses each visit HBP clients an average of 4 times (8 times combined), with the range being 1 to 18 visits for nurses and 1 to 9 visits for dietitians (1 to 24 visits for nurses and dietitians combined).

Comparison of Treatment Groups

We compared Groups A and B for each initially assessed risk factor to determine which, if any, risk factors may confound with the number of visits. In other words, did the percentage of clients differ from the 50% we would theoretically expect? We found that Group A clients, who received the lower number of visits, differed significantly from those in Group B in only three of 24 risk factors - smoking, street drugs and diet.

For two of those risk factors (smoking and street drugs), there was a greater percentage of clients in Group A than Group B. Both these risk factors are known to be associated with LBW. Therefore, with higher percentages of clients with these risk factors receiving fewer number of visits, it may be that smoking or the use of street drugs have the greater influence than the lower exposure to HBP on LBW. Of the 78 clients assessed with smoking as a risk factor, there were 8 (10.3%) LBW babies born, 5 to mothers in Group A and 3 to mothers in Group B. Of the 19 clients assessed with use of street drugs, 3 (15.8%) had LBW infants. All three were born to mothers in Group A. Because of the small sample size, we cannot confidently conclude that these are confounding variables. Further study of these possible confounding variables would be appropriate when a larger sample size is available.

The other risk factor with statistically significant differences between the treatment groups was inadequate diet. For 296 clients assessed with inadequate diet, 46% were in Group A and 54% in Group B. There were 22 (7.4%) LBW babies to mothers with inadequate diets, 15 of them in Group A and 7 in Group B. Therefore, differences in LBW outcomes can not be explained by the variance in membership in Group A and Group B for mothers initially assessed with inadequate diet.

Treatment groups were also compared by the number of initially assessed risk factors. Thirty percent of HBP clients who completed the program and had singleton births were assessed with 1 to 3 risk factors, 52% had 4 to 6 risk factors, and 18% had 7 or more risk factors. Clients from these 3 levels of risk were evenly distributed among Group A and Group B.

Analysis

We have assumed from the above comparisons that Groups A and B, while not exactly the same in terms of initial risk factors, were similar enough that differences in their outcomes likely reflect changes due to exposure to the HBP program, not baseline differences. We therefore continued our analysis in 2 areas:

1. Comparison of groups A and B with respect to behavioural, health and socioenvironmental change and LBW, and

2. Projection of outcomes for a hypothetically reduced or expanded program, by number of visits per client.

The key strategic areas for intervention in HBP that are included in this study are: food group deficits, initial weight loss or poor weight gain, smoking, continuous high stress, depression and financial problems. These intervention areas target behavioural, health and socioenvironmental change in selected risks and conditions.

For all above intervention areas, the treatment groups were compared according to improvement in 1) selected risks and conditions, and 2) incidence of low birthweight. Improvement means:

- for food groups - change of intake from a deficit to regular consumption of the recommended number of servings;

- for initial weight loss or poor weight gain - significant weight gain as assessed by a professional dietitian;

- for smoking - cutting down or quitting, as reported by client;

- continuous high stress, depression, financial - a significant improvement as assessed by a professional nurse;

- incidence of low birthweight - percentage of mothers with babies weighing less than 2500 grams at birth.

We compared incidence of LBW for Groups A and B for all HBP clients and for the clients who received intervention for selected risks or conditions. Next we compared the percentages of clients who showed improvements in those same risks and conditions by treatment group. This process allows us to determine if clients in Group B had lower incidence of LBW and greater incidence of improvement than Group A.

To estimate the number of LBW babies prevented by the HBP program, we determined what the LBW rate would have been if all clients had received only 1-4 visits or 1-8 visits. We also projected the LBW rate of those who received more than 8 visits on to the entire HBP population, to estimate what the program could potentially accomplish with an optimal number of visits.

RESULTS

The overall low birthweight rate of all clients in the study (singleton births only) was 6.5% (n=27). This approaches the rate for the general population in Toronto of 5.1%.

Figure 2 shows that as the number of home visits increased, the percentage of LBW babies decreased from 27.7% for clients who received only 1-2 visits, to 10.4% of those who received 3-4 visits. The high rate of 27.7% LBW among clients who received very few home visits is similar to findings in a previous HBP evaluation in 1990. The percentage of LBW dips below the City of Toronto average rate of 5.1% at 5-6 visits, and remains close to that level except for those who received 13-14 visits. This latter increase may be because these mothers are at highest risk.

When comparing birthweight outcomes between the two treatment groups who received intervention for the five key areas, we found substantial differences (Figure 3). Of all mothers in Group A, 7.7% gave birth to LBW babies, while only 4.8% of those in Group B, with the higher number of visits, had LBW babies. Figure 3 shows the trend towards fewer LBW babies among mothers in Group B who received more visits. The only key area of intervention that does not follow this pattern is depression, where the two groups were almost equal.

Figure 4 compares groups according to improvement in the five key intervention areas. It illustrates that improvements are greater in Group B, the mothers who received more visits. Only continuous high stress shows an opposite trend.

Based on the study analysis, Table 4 (Appendix) summarizes the projected outcomes for hypothetical situations in which all clients receive more or fewer visits compared with actual program outcomes. This assumes that the incidence of LBW remains the same for each category of visits. Because clients who received 1 - 4 visits had a 14.3% LBW, we project that if all our clients had only 1-4 visits, 59 LBW babies would have been born over the same two year period, or 32 more than actual. Similarly, if all clients received a maximum of 8 visits, we project that with a LBW rate of 7.7%, 32 LBW babies would have been born over the same 2 year period, only 5 more than the actual number. Projecting further, if we were to provide more than eight visits to all HBP clients, the number of LBW babies would be 7 less than actual, assuming that the LBW percentage would remain at 5.1% for clients who received more than eight visits. This would mean only 20 babies weighing less than 2500 grams would be born over a 2 year period to an equivalent group of about 413 high risk pregnant women who received HBP intervention. This incidence of LBW is about the same as for the City of Toronto as a whole.

While these predictions are conjecture, it is still probable that the HBP program likely prevented between 5 and 32 low birthweight babies during the study period. In the absence of an adequate control group, it is not possible to predict exactly the number of LBW babies prevented because of the HBP program.

In summary, the key findings from this study are as follows:

1. The incidence of low birthweight among HBP clients between June 1994 and June 1996 was 6.5%. Among HBP clients who received only one or two visits, the incidence of LBW is 27.7%, compared to 5.1% for clients who received more than eight visits. During the same period, the incidence of LBW for the City of Toronto was approximately 5.1%.

2. HBP clients who received a greater number of visits (Group B) had fewer LBW infants than those who received less visits (Group A), without controlling for improvement. LBW for mothers in Group A is 7.7% and for Group B it is 5.1%.

3. Improvements in behavioural, health and socioenvironmental factors are positively associated with the number of home visits by dietitians and public health nurses. Clients who received a greater number of visits (Group B) had higher rates of improvement than those with fewer visits (Group A) for each risk or condition except continuous high stress.

4. The estimated number of LBW babies prevented by the HBP program is substantial, and the program could potentially have an even greater impact if women entered the program earlier and maintained participation to the end of their term. This would allow the time necessary for a minimum of eight visits, the number projected to ensure optimal effect.

CONCLUSION

REFERENCES

1. Health Canada. Strengthening Prenatal Health Promotion for Disadvantaged Families. Ottawa, 1994.

2. The Canadian Institute of Child Health. Prevention of Low Birthweight in Canada: Literature Review and Strategies. Toronto: Health Promotion Branch, Ontario Ministry of Health, 1993.

3. Canadian Institute of Child Health. Prevention of Low Birth Weight in Canada: Literature Review and Strategies. 1992.

4. Ciliska, D., Hayward, S., Thomas, H., Mitchell, A., Dobbins, M., Underwood, J., Rafael, A., and Martin, E. The Effectiveness of Home Visiting as a Delivery Strategy for Public Health Nursing Interventions - A Systematic Overview. Quality of Nursing Worklife Research Unit, McMaster University - University of Toronto. Working Paper Series 94-7. November, 1994.

Table 1

Incidence of low birthweight for key intervention areas

Table 2

Clients showing improvement in key intervention areas

* Improvement means: For food groups - change of intake from a deficit to usual consumption of the recommended number of servings; for initial weight loss or poor weight gain - significant weight gain as assessed by a professional dietitian; for smoking - cutting down or quitting; continuous high stress, depression, financial - a significant improvement as assessed by a professional nurse.

Table 3

Incidence of low birthweight among clients in treatment groups, for key intervention areas

Table 4

Comparison of LBW outcomes for actually and hypothetically reduced or expanded

versions of the HPB program